Tracing Enriched Mantle Components along the Gakkel Ridge, Arctic
Ocean
A Thesis Presented to
the Faculty of the Department of Earth and Atmospheric Sciences
University of Houston
in Partial Fulfillment
of the Requirements for the Degree
Master of Science
By
Nam Hien Nguyen
December, 2013
Tracing Enriched Mantle Components along the Gakkel Ridge, Arctic
Ocean
Nam Hien Nguyen
APPROVED:
Supervisor, Dr. Jonathan Snow
Dr. Alan Brandon
Dr. Yongjun Gao
Dr. Steven Bergman
Dean, College of Natural Sciences and Mathematics
ii
ACKNOWLEDGEMENTS
Thank you, Dr. Snow, for all the help.
December, 2013
iii
Tracing Enriched Mantle Components along the Gakkel Ridge, Arctic
Ocean
An Abstract of a Thesis
Presented to
the Faculty of the Department of Earth and Atmospheric Sciences
University of Houston
in Partial Fulfillment
of the Requirements for the Degree
Master of Science
By
Nam Hien Nguyen
December, 2013
iv
ABSTRACT
Mid-ocean ridge basalts (MORBs) along the Gakkel Ridge in the Arctic Eurasian basin
have varying levels of trace element enrichment. The purpose of this research is to determine the
extent to which the trace element enrichment of basalts along the Gakkel Ridge is influenced by
magmatic differentiation processes or mantle heterogeneity. This was accomplished by
calculating the effects of magmatic differentiation on basalt trace element ratios and by defining
the compositions of mantle components in the mantle along Gakkel Ridge and interpreting their
origins, scales, and distribution. Electron microprobe and laser-ablation inductively-coupledplasma mass-spectrometer (LA-ICP-MS) instruments were used to analyze major element and
trace element concentration data in glass chips of basalt samples dredged from the Gakkel Ridge.
As a consequence of Gakkel Ridge’s ultraslow spreading rate, fractional crystallization has only a
minimal effect on trace element ratios of Gakkel Ridge MORBs. Trace elements plotted on
ternary plots reveal four main aspects about the nature of the enriched component along Gakkel
Ridge: 1) Sparsely Magmatic Zone (SMZ) and Eastern Volcanic Zone (EVZ) basalt trace element
data plot along a two-component mixing line trending between a depleted MORB mantle (DMM)
composition and an average Mid-Atlantic Ridge enriched-MORB (MAR E-MORB) composition.
This mixing line is separate from the mixing line on which the Western Volcanic Zone (WVZ)
trace element data plot, which trends between the DMM and the high-K2O Central Lena Trough
(CLT) source; 2) The level of enrichment in MORB samples of the SMZ and EVZ varies from
being relatively similar across a span of 97 km in the EVZ at 31-47°E, to changing significantly
between individual dredging stations in the SMZ at 13°E, 19°E, and in the EVZ at 55°E. This
shows that the size and distribution of enriched reservoirs drastically vary; 3) Trace elements
indicate that the main enriched source of the EVZ and SMZ basalts resemble MAR E-MORB
source mantle and is not continental in origin; and 4) At least one additional enriched source
exists along Gakkel Ridge, as evidenced by the chemical signature in the EVZ at 55°E, which
shows at least three different mixing component compositions.
v
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ............................................................................................... iii
ABSTRACT .........................................................................................................................v
1) INTRODUCTION ...........................................................................................................1
1.1) Geochemical connection between mantle and MORB .............................................1
1.2) Basalt enrichment......................................................................................................5
1.3) Trace element ratios – ternary plots ..........................................................................7
1.4) Magma differentiation.............................................................................................15
1.4.1) Fractional crystallization .................................................................................15
1.4.2) Partial melting .................................................................................................23
1.4.3) Spreading rate effects.......................................................................................24
1.5) Mantle heterogeneity...............................................................................................26
1.5.1) Recycling ..........................................................................................................26
1.5.2) Veins .................................................................................................................28
1.5.3) Magmatic mixing..............................................................................................29
1.5.4) OIB mantle components ...................................................................................30
2) SETTING.......................................................................................................................34
2.1) Magmato-tectonic zones .........................................................................................36
2.1.1) Western Volcanic Zone (WVZ) (7°W - 3°E) .....................................................38
2.1.2) Sparsely Magmatic Zone (SMZ) (3° - 29°E) ....................................................38
2.1.3) Eastern Volcanic Zone (EVZ) (29°E - 85°E) ...................................................39
2.2) Adjacent Arctic Basin features ...............................................................................40
3) BACKGROUND ...........................................................................................................40
3.1) AMORE 2001 .........................................................................................................41
3.2) AGAVE 2007 ..........................................................................................................48
3.3) Earlier studies..........................................................................................................49
3.3.1) Geochemistry of Gakkel Ridge MORBs ...........................................................49
3.3.2) Geochemistry of other Arctic Basin basalts .....................................................50
4) METHODS ....................................................................................................................53
vi
4.1) LA-ICP-MS.............................................................................................................54
4.2) Electron microprobe................................................................................................55
4.3) Uncertainty ..............................................................................................................55
4.4) Two-component mixing lines .................................................................................60
5) RESULTS ......................................................................................................................60
5.1) Data .........................................................................................................................62
5.1.1) Gakkel Ridge major element compositions ......................................................63
5.1.2) Gakkel Ridge trace element compositions .......................................................64
5.2) Ternary plots – Arctic MORBs ...............................................................................66
6) DISCUSSION ................................................................................................................69
6.1) Enriched component reservoirs...............................................................................69
6.1.1) Enriched reservoirs: size & distribution ..........................................................71
6.1.2) Enriched component: origin ............................................................................78
6.2) Isotopic boundary....................................................................................................84
7) CONCLUSIONS ...........................................................................................................84
8) APPENDIX ...................................................................................................................87
8.1) Appendix 1: XY plots .............................................................................................87
8.2) Appendix 2: Global MORB ternary plots ...............................................................97
8.3) Appendix 3: NbKL & DSL ternary plots ..............................................................100
8.4) Appendix 4: NKL ternary plots - mixing lines .....................................................107
8.5) Appendix 5: Ternary plots - enriched continental magmas ..................................111
8.6) Appendix 6: NKL & TKL ternary plots - Gakkel Ridge volcanic centers ...........117
8.7) Appendix 7: Tables - basalt standards ..................................................................126
8.8) Appendix 8: Tables - relative uncertainty ranges .................................................130
9) REFERENCES ............................................................................................................131
LIST OF FIGURES
Figure 1a. Diagram - MORB subaxial melting regime triangle - generalized.....................4
Figure 1b. Diagram - MORB subaxial melting regime triangle - ultraslow ........................4
Figure 2. Spidergram - N-MORB, T-MORB, & E-MORB .................................................5
vii
Figure 3a. Ternary plots - global MORBs - NKL..............................................................10
Figure 3b. Ternary plots - global MORBs - TKL ..............................................................11
Figure 3c. Ternary plots - global MORBs - YSL ..............................................................12
Figure 4a. Ternary plots - fractional crystallization – olivine - YSL.................................19
Figure 4b. Ternary plots - fractional crystallization – olivine - TKL ................................20
Figure 4c. Ternary plots - fractional crystallization – clinopyroxene - YSL .....................21
Figure 4d. Ternary plots - fractional crystallization – clinopyroxene - TKL ....................22
Figure 4e. Ternary plots - fractional crystallization – plagioclase - YSL..........................23
Figure 5a. Ternary plots – OIB mantle components - NKL ..............................................32
Figure 5b. Ternary plots – OIB mantle components - TKL ..............................................33
Figure 5c. Ternary plots – OIB mantle components - YSL ...............................................34
Figure 6. Map - Arctic Basin .............................................................................................35
Figure 7. Map - Gakkel Ridge AMORE 2001 sample locations ................................. 43-47
Figure 8. Map - Gakkel Ridge AGAVE 2007 sample locations........................................48
Figure 9. XY plots – Radiogenic isotopes of Arctic MORBs............................................52
Figure 10. Ternary plot – uncertainty polygons of samples 199-1-2 & 252-1 ..................59
Figure 11. Spidergram - Arctic MORBs ............................................................................61
Figure 12a. Ternary plots - Arctic MORBs - NKL............................................................66
Figure 12b. Ternary plots - Arctic MORBs - TKL ............................................................67
Figure 12c. Ternary plots - Arctic MORBs - YSL ............................................................68
Figure 13a. Ternary plots – Gakkel MORB uncertainty polygons – SMZ13 ....................73
Figure 13b. Ternary plots – Gakkel MORB uncertainty polygons – SMZ19 ...................74
Figure 13c. Ternary plots – Gakkel MORB uncertainty polygons – EVZ31-37-43 .........75
Figure 13d. Ternary plots – Gakkel MORB uncertainty polygons – EVZ55 ....................76
Figure 14a. Ternary plots - mixing lines - TKL - EM2 .....................................................79
Figure 14b. Ternary plots - mixing lines - TKL - EM1 .....................................................80
Figure 14c. Ternary plots - mixing lines - TKL - E-MORB ..............................................81
Figure A1a. XY plots - Ba/TiO2 vs. MgO. ........................................................................88
Figure A1b. XY plots - Dy/Yb vs. MgO ...........................................................................88
Figure A1c. XY plots - K2O/La vs. MgO ..........................................................................89
viii
Figure A1d. XY plots - K2O/TiO2 vs. MgO ......................................................................89
Figure A1e. XY plots - La/SmA1e vs. MgO .....................................................................90
Figure A1f. XY plots - Nb/La vs. MgO .............................................................................90
Figure A1g. XY plots - Ba/TiO2 vs. Mg# ..........................................................................91
Figure A1h. XY plots - Dy/Yb vs. Mg# ............................................................................91
Figure A1i. XY plots - K2O/La vs. Mg# ............................................................................92
Figure A1j. XY plots - K2O/TiO2 vs. Mg# ........................................................................92
Figure A1k. XY plots - La/Sm vs. Mg#.............................................................................93
Figure A1l. XY plots - Nb/La vs. Mg# ..............................................................................93
Figure A1m. XY plots - Ba/TiO2 vs. Longitude ................................................................94
Figure A1n. XY plots - Dy/Yb vs. Longitude ...................................................................94
Figure A1o. XY plots - K2O/La vs. Longitude ..................................................................95
Figure A1p. XY plots - K2O/TiO2 vs. Longitude ..............................................................95
Figure A1q. XY plots - La/Sm vs. Longitude ....................................................................96
Figure A1r. XY plots - Nb/La vs. Longitude .....................................................................96
Figure A2a. Ternary plots - global MORBs - NKL ...........................................................97
Figure A2b. Ternary plots - global MORBs - TKL ...........................................................98
Figure A2c. Ternary plots - global MORBs - YSL ...........................................................99
Figure A3a. Ternary plots - NbKL - global & enriched MORBs ....................................101
Figure A3b. Ternary plots - NbKL - OIB mantle components ........................................102
Figure A3c. Ternary plots - NbKL - global & Arctic MORBs ........................................103
Figure A3d. Ternary plots - DSL - global & enriched MORBs ......................................104
Figure A3e. Ternary plots - DSL - OIB mantle components ...........................................105
Figure A3f. Ternary plots - DSL - global and Arctic MORBs ........................................106
Figure A4a. Ternary plots - mixing lines - NKL - EM2 ..................................................108
Figure A4b. Ternary plots - mixing lines - NKL - EM1 ..................................................109
Figure A4c. Ternary plots - mixing lines - NKL - E-MORB ..........................................110
Figure A5a. Ternary plots - enriched components - NKL ...............................................112
Figure A5b. Ternary plots - enriched components - TKL ...............................................113
Figure A5c. Ternary plots - enriched components - YSL................................................114
ix
Figure A5d. Ternary plots - enriched components - NbKL .............................................115
Figure A5e. Ternary plots - enriched components - DSL................................................116
Figure A6a. Ternary plots - NKL - SMZ13 .....................................................................118
Figure A6b. Ternary plots - NKL - SMZ19 .....................................................................119
Figure A6c. Ternary plots - NKL - EVZ31-EVZ37-EVZ-43 ..........................................120
Figure A6d. Ternary plots - NKL - EVZ55 .....................................................................121
Figure A6e. Ternary plots - TKL - SMZ13 .....................................................................122
Figure A6f. Ternary plots - TKL - SMZ19 ......................................................................123
Figure A6g. Ternary plots - TKL - EVZ31-EVZ37-EVZ43 ...........................................124
Figure A6h. Ternary plots - TKL - EVZ55......................................................................125
LIST OF TABLES
Table 1. DMM, MAR N-MORB, & MAR E-MORB trace element compositions .............6
Table 2. Uncertainty calculations of SMZ19 samples 199-1-2 & 252-1. ..........................57
Table 3. Gakkel Ridge major element compositions .........................................................63
Table 4. Gakkel Ridge trace element compositions...........................................................64
Table A7a. Standard analyses - major elements - VG-2 ..................................................127
Table A7b. Standard analyses - trace elements - BIR-1 (part 1) .....................................127
Table A7c. Standard analyses - trace elements - BIR-1 (part 2) .....................................127
Table A7d. Standard analyses - trace elements - KL2-G (part 1) ....................................128
Table A7e. Standard analyses - trace elements - KL2-G (part 2) ....................................129
Table A8a. Relative uncertainty ranges - major elements ...............................................130
Table A8b. Relative uncertainty ranges - trace elements (part 1)....................................130
Table A8c. Relative uncertainty ranges - trace elements (part 2) ....................................130
x
1)
INTRODUCTION
1.1)
Geochemical connection between mantle and MORB
Earth’s mantle is defined as the region in the planet’s interior located between the
crust and the core. The mantle shares an important geochemical connection with the
crust, because the crust was originally formed from melts derived from the mantle; thus,
the geochemistry of crustal rocks reflects their genetic relationship with the rocks of the
mantle (Hofmann, 1988). Indeed, it can be said that the crust is “genetically” related to
the mantle.
To understand how the crust and the mantle are geochemically related, it is
important to understand the processes by which crustal rocks are formed. At mid-ocean
ridges, seafloor spreading separates the tectonic plates, leading to upwelling of
asthenospheric mantle material beneath the ridge axes (Langmuir et al., 1992; Hofmann,
1997; Klein, 2003b). As mantle convection brings mantle rocks upward, the pressure
exerted downwards by the overlying rocks upon mantle rocks decreases, resulting in
decompression melting of the rising mantle rocks, thereby producing magma melts that
flow upwards, erupt, and cool to form the oceanic crust, the upper layer of which is
composed of mid-ocean ridge basalt (MORB) (Fig. 1; Langmuir et al., 1992; Klein,
2003b).
Peridotite is a rock type mainly composed of olivine, pyroxene, garnet, and spinel
group minerals. Peridotite is the dominant rock-type of the upper mantle (Helffrich &
Wood, 2001; Sobolev et al., 2005). The peridotite-dominated region extends down at
least as deep in the mantle as 150 km (Helffrich & Wood, 2001). This region of the
mantle is relevant to the study of MORB chemistry, because MORBs are derived from
1
shallow mantle depths, meaning that MORBs are derived primarily from peridotites. The
mineral olivine, which composes more than 50% of earth’s upper mantle, is the most
abundant mineral in the peridotite portion of the mantle (Sobolev, 2005).
The depleted MORB mantle (DMM) is recognized as the most widespread
magmatic source component in the upper mantle and the dominant source of magmatic
melts in the generation of mid-ocean ridge magma (Zindler & Hart, 1986). Basalts
derived from this source are characterized by a depletion in the more highly incompatible
elements, such as light rare earth elements, relative to the primordial undifferentiated
primitive mantle (PM) composition (Zindler & Hart, 1986). The depletion of
incompatible elements in DMM is considered to be evidence that the DMM has
experienced previous partial melting events, during which the more incompatible
elements were extracted in greater proportion than the less incompatible elements
(Zindler & Hart, 1986).
The degree of enrichment in MORBs can vary with distance along a single ridge.
For instance, a notable example of such a ridge is the ultraslow spreading Gakkel Ridge
in the Arctic Basin, where the MORBs show significantly different degrees of variation
in isotopic and trace element enrichment (Michael et al., 2003; Goldstein et al., 2008;
Shaw et al., 2010).
It has generally been assumed that two major factors can affect basalt
compositions (e.g., Grove et al., 1992; Klein et al., 2003b): 1) magmatic differentiation
processes experienced by magmas prior to erupting (e.g., Ohara, 1965; Kay et al., 1970)
and/or 2) chemical and mineralogical compositional heterogeneity in the mantle source
from which magmas were derived (e.g., Schilling, 1973).
2
This study examines how the chemistry of Gakkel Ridge MORB compositions
were affected by differentiation processes and by mantle heterogeneity. A better
understanding of Gakkel Ridge chemical diversity can lead to better understandings of
slow spreading ridge evolution as well as the global MORB diversity.
3
Figure 1a. Generalized MORB subaxial melting regime triangle, showing varying
percentages of mantle melting and melt extraction correlating with the depth and
direction of the melt flow (Modified from Langmuir & Forsyth, 2007).
Figure 1b. MORB subaxial melting triangle at ultraslow spreading ridges (Modified
from Langmuir & Forsyth, 2007)
4
1.2)
Basalt enrichment
Basalt geochemistry studies generally categorize MORBs based on their relative
levels of incompatible trace element enrichment or depletion. Under this classification
system, MORBs vary in composition from normal mid-ocean ridge basalt (N-MORB) to
trace element enriched mid-ocean ridge basalt (E-MORB). E-MORBs are characterized
by elevated amounts of incompatible trace elements compared to N-MORB. Intermediate
compositions are categorized as transitional mid-ocean ridge basalt (T-MORB) (Sun &
McDonough, 1989).
Figure 2. Spidergram showing average trace elements of Mid-Atlantic Ridge (MAR) NMORB, T-MORB, and E-MORB (Klein, 2003b). Average Macquarie Island basalt
(Kamenetsky et al., 2000) and Southwest Indian Ridge - oblique segment MORBs
(Standish et al., 2008) are plotted as additional examples of different E-MORB
compositions. Elements are normalized to pyrolite (McDonough & Sun, 1995).
5
The La/Sm ratio, which has long been used to measure enrichment in MORBs
(e.g., Schilling, 1971), compares the highly incompatible rare earth element lanthanum
(La) with the less incompatible rare earth element samarium (Sm). In order to distinguish
between N-MORB, T-MORB, and E-MORB, the La/Sm ratio can be used to define the
boundaries in the levels of enrichment between the different MORB types (Sun &
McDonough, 1989). However, the ratios corresponding with each type of MORB are not
precisely defined by any widely agreed upon convention, so they must be arbitrarily
decided for this study. In this thesis, the La/Sm ratio that was used to define the MORB
types are: N-MORB La/Sm <0.8; T-MORB La/Sm 0.8-1.8; E-MORB La/Sm >1.8.
Trace element concentrations of average MAR N-MORB and E-MORB, as well
as a DMM composition, are shown in Table 1 for comparison with Gakkel Ridge
concentration data. Among these compositions, DMM is the most depleted, MAR NMORB is the next most depleted, and MAR E-MORB is the most enriched.
Rb
DMM
MAR N-MORB
MAR E-MORB
Sr
0.1
0.4
13
Tb
DMM
MAR N-MORB
MAR E-MORB
Y
0.1
104
221
Dy
0.08
0.58
0.73
0.53
4.07
4.94
Zr
4.1
24.0
27.4
Nb
7.9
57
121
Cs
0.2
1.1
20
Ho
Er
Tm
Yb
0.122
0.37
0.060
0.892
2.62
0.377
1.06
3.08
0.432
Ba
0.001
0.005
0.128
0.4
2.7
3.0
La
1.2
4.6
146
Ce
0.2
1.8
13.5
Pr
0.8
6.0
29.2
Nd
0.1
1.1
3.8
0.7
6.3
16.8
Sm
Eu
Gd
0.27
0.11
0.40
2.37
0.94
3.28
4.22
1.44
4.77
Lu
Hf
Ta
Pb
Th
U
TiO2
Sc
0.063
0.20 0.014
0.02
0.01 0.005
4.49
16.3
0.377
1.62 0.055
0.27
0.05 0.022
32.1
38.7
0.428
2.25 0.347
1.1
0.38 0.127
41.5
40.3
Table 1. DMM composition calculated by Salters and Stracke (2004). Average MAR NMORB and E-MORB data compiled from PetDB (Lehnert et al., 2000). TiO2 is reported
in weight percent. All other concentrations are reported in ppm. N-MORB is defined as
basalts (glass samples) having a La/Sm ratio of less than 0.8. E-MORB is defined as
basalts (glass samples) having a La/Sm ratio of greater than 1.8.
It should be noted, however, that all E-MORBs are not compositionally identical,
but instead, different suites of E-MORBs have been recognized (Fig. 2). For example,
6
Macquarie Island basalts (Fig. 2; Fig. 3; Kamenetsky et al., 2000) and Southwest Indian
Ridge (SWIR) oblique segment basalts (Fig. 2; Fig. 3; Standish et al., 2008) can be
considered to be E-MORB compositions, but their trace element patterns are also distinct
from each other and from the average Mid-Atlantic Ridge (MAR) E-MORB. The
distinctions between different enriched MORBs may indicate that all E-MORBs do not
share the same reason for being enriched.
1.3)
Trace element ratios - ternary plots
Ternary plots are triangular-shaped graphs on which data from three different
variables can be plotted. Such plots are especially useful for comparing trace element
concentration ratios, because they simultaneously illustrate relative proportions of three
different trace elements.
Trace element ratios are useful in the study of MORB compositions because
certain element ratios can remain relatively constant throughout the process of mantle
melt extraction, even after the magmatic melts have experienced differentiation processes
(Salters & Stracke, 2004). This resistance to change, which is especially apparent at low
degrees of differentiation, means that the ratios in basalts closely reflect the original
ratios in the mantle. For instance, La/Sm is one example of a ratio that is relatively
insensitive to fractional crystallization (Holness & Richter, 1982).
Three sets of three trace elements each were chosen for the ternary plots
throughout this thesis: Na2O & K2O & La (NKL), TiO2 & K2O & La (TKL), and Yb &
Sm & La (YSL). The relative incompatibilities of these elements are: K>La>Sm>Ti>Yb
(Sun & McDonough, 1989). Na2O appears to be more compatible than K2O and La and
7
similar in incompatibility to Sm (e.g., Fig. 3a). These elements were chosen for their
variety in different relative levels of incompatibility and for some of their unique
characteristics.
A unique characteristic of the rare earth elements (e.g., La, Sm, and Yb) is that
they share similar chemical and physical properties, with the primary difference being
that their ionic radii decrease with increasing atomic number, so the size of these atoms
are the main factor determining differences in their differentiation behavior. The YSL
ternary plot features only rare earth elements, ranging from one of the largest rare earth
elements, La, to the intermediate-sized Sm, to one of the smallest rare earth elements, Yb.
La is especially useful for comparisons between elements, because it is the most
incompatible of the rare earth elements, which is why La is included in all three sets of
elements.
K2O is featured on both the TKL and NKL ternary plots primarily so that basalts
from Gakkel Ridge can be compared to other basalts of the Arctic Basin, because many
of the basalts of the Arctic Basin, notably those of central Lena Trough, feature unusually
enriched K2O compositions (Nauret, 2011).
One of the useful applications of the ternary plots is that they allow distinction of
samples derived from more enriched sources from samples derived from more depleted
sources. On each ternary plot, the trend between the depleted and enriched sources is
easily identifiable, because it tends to trend between the average N-MORB and E-MORB
points. The depleted to enriched trend also tends to correlate with the DMM to PM trend,
although PM is not as enriched as the average E-MORB. Overall, this means that the
samples from more depleted sources plot closer to the average N-MORB and depleted
8
MORB mantle (DMM) points, whereas the samples from more enriched sources plot
closer to the average E-MORB points.
To illustrate the global variation in levels of enrichment in MORBs, Figure 3
presents the NKL, TKL, and YSL trace element ternary plots depicting relative trace
element compositions in global MORB and enriched basalt datasets, along with points
representing average N-MORB, T-MORB, and E-MORB from different ocean ridges.
9
Figure 3a. NKL Ternary plot showing relative concentrations of Na2O (wt.% x 5), K2O
(wt% x 30), & La (ppm) in basalt glass samples from the Mid-Atlantic Ridge (MAR),
East Pacific Rise (EPR), Southwest Indian Ridge (SWIR), Southeast Indian Ridge (SEIR)
(PetDB; Lehnert et al., 2000), and Southwest Indian Ridge – Oblique segment (Standish
et al., 2008), and Macquarie Island (Kamenetsky et al., 2000). Average N-MORB, TMORB, and E-MORB points for the MAR and SWIR are calculated based on the La/Sm
ratio, with N-MORB La/Sm <0.8; T-MORB La/Sm 0.8-1.8; E-MORB La/Sm >1.8. Also
plotted as reference points on these ternary plots are the depleted MORB mantle (DMM)
(Salters & Stracke, 2004) and primitive mantle (PM) (Sun & McDonough, 1989),
although a Na2O concentration for PM was not found, so PM does not appear in this
figure, but does appear in the TKL (Fig. 3b) and YSL (Fig. 3c) ternary plots.
10
Figure 3b. TLK Ternary plot showing relative concentrations of TiO2 (wt.% x 5), K2O
(wt.% x 30), & La (ppm) in MORB glass samples. Data are from the same references as
those in Fig. 3a.
11
Figure 3c. YSL Ternary plot showing relative concentrations of Yb (ppm), Sm (ppm x
2), & La (ppm) in MORB glass samples. Data are from the same references as those in
Figure 3a. Strangely, the DMM composition (Salters & Stracke, 2004) plots far off of the
MORB array on the YSL triangle, even though it should plot close to average N-MORB
as it does on the NKL (Fig. 3a) and TKL (Fig. 3b) triangles.
The NKL, TKL, and YSL ternary plots of Figure 3 all show oceanic basalts
forming a single array between enriched and depleted endmembers in composition space.
On the YSL ternary plots (i.e., Fig. 3c), the estimated DMM composition (Salters
& Stracke, 2004) plots significantly far off of the main MORB array on the YSL triangle
12
(e.g., Fig. 3c). This position is unexpected, because by definition, N-MORB is the
melting product of DMM, so DMM should always plot close to average N-MORB, which
is indeed how it appears on the NKL and TKL triangles. Also, these DMM compositions
are not only inconsistent with the data from this study, but also with the global MORB
composition data obtained from PetDB (Lehnert et al., 2000). An alternate estimated
DMM composition estimated by Workman & Hart (2005), different from the
composition from the estimate by Salters & Stracke (2004), was also plotted on the
ternary plots as a test to see whether the same problem persists. However, much like the
DMM from Salters & Stracke (2004), the alternate DMM also plotted off of the MORB
array, significantly far away from average the average N-MORB point. The reason that
the DMM composition estimated by Salters & Stracke (2004) and by Workman & Hart
(2005) both appear to be wrong for the YSL triangle is unknown.
One possible explanation is that unlike the NKL triangle (e.g., Fig. 3a) and TKL
triangle (e.g., Fig. 4b), which include oxides (i.e., Na 2O, TiO2, K2O), the YSL triangle
features only rare earth elements (i.e., Yb, La, Sm), so it may be possible that something
is wrong with the rare earth element estimates in the DMM composition by Salters &
Stracke (2004). To investigate this possibility, Salters & Stracke’s DMM composition
concentrations of the rare earth elements Eu, Gd, Tb, Dy, Ho, Er, Tm, and Yb (calculated
as a percentage relative to La and Sm, i.e., the percentage of the aforementioned element
added to the percentages of La and Sm sum to 100%) were compared to the
concentrations of those elements in the average MAR N-MORB composition. Those
particular rare earth elements were chosen because Salters & Stracke (2004) used the
same constraining ratio denominator in their calculations for all of those elements, but not
13
for the rest of the rare earth elements. The amount of deviation was calculated by
comparing the compositions of those elements in DMM and MAR N-MORB as
percentages relative to La and Sm (i.e., when the percentage of that element is added to
the percentages of La and Sm, it sums to 100%) and then comparing the percent
difference between the DMM concentration and the MAR N-MORB concentration of
that element. The results of these calculations show that Eu, Gd, Tb, Dy, Ho, Er, Tm, and
Yb in Salters and Stracke’s DMM composition differs when compared to those
compositions in the average MAR N-MORB by 4.8%, 0.7%, 1.1%, 4.5%, 8.0%, 10.6%,
20.7%, and 14.5%, respectively. There appears to be a trend of increasing deviation from
the lighter rare earth elements (i.e., Eu, Gd, Tb, Dy) to the heavier rare earth elements
(i.e., Ho, Er, Tm, Yb). The exact cause of the heavier rare earth elements being less
correlated with MAR N-MORB (and thus, with the MORB array) is unknown, but the
amount of discrepancy in Salters & Stracke’s (2004) DMM compositions appears to
increase in correlation with the gradual lanthanide contraction from light rare earth
elements to heavy rare earth elements.
14
1.4)
Magma differentiation
One idea that is often discussed by studies of MORB geochemistry is that the
diversity observed in global MORB compositions reflects an influence from two main
factors: 1) magmatic differentiation processes and/or 2) mantle heterogeneity (Klein,
2003b). The extent to which each factor influences the MORB composition is still a
matter of debate, but one widely held belief is that among the many processes causing
chemical differentiation in magmas, the most dominant processes that can affect the
composition of MORBs are fractional crystallization and partial melting (e.g., Grove et
al., 1992; Klein, 2003b). Between crystallization and melting, crystallization is often
considered to have a greater impact on MORB compositions (e.g., Klein, 2003b), so this
study devotes more attention to fractional crystallization.
1.4.1) Fractional crystallization
Fractional crystallization is the process of formation and removal of mineral
precipitates from the magmatic melt, during cooling of the magmatic melt, which occurs
at shallow levels in the uppermost mantle and crust (e.g., Grove et al., 1992; Klein,
2003b). By incorporating elements from the magmatic melt into the crystallizing solid
mineral precipitates, the fractional crystallization process changes the chemical
composition of the uncrystallized liquid magma (e.g., Grove et al., 1992; Klein, 2003b).
Basalts erupted from the same sub-axial magma reservoir may be derived from
varying stages of crystallization, resulting in MORBs with different chemical
compositions (Klein, 2003b). However, because the effects of fractional crystallization
are well understood – to a better degree than any of the other magma differentiation
15
processes – the chemical signature of fractional crystallization can be distinguished from
chemical signatures of other magma differentiation processes and variations in source
composition (Fig. 4, Klein, 2003b). In general, fractional crystallization is only capable
of causing relatively minor enrichments in incompatible elements in basalts (Hofmann,
1988).
The extent of the effect that the fractional crystallization process has on the
concentrations of trace elements in magmatic melts can be calculated using the fractional
crystallization equation:
Cl/Co=(1-X)D-1
with Co denoting the concentration in the original liquid, C l denoting the concentration in
the remaining liquid, X denoting the fraction of the material crystallized, and D denoting
the partition coefficient. The fractional crystallization equation was obtained from White
(2013).
Figures 4a-4e show the results of fractional crystallization of olivine in a DMM
(Salters & Stracke, 2004) magma composition. Fractional crystallization calculations
were conducted for the mineral phases olivine (Fig. 4a-4b), clinopyroxene (Fig. 4c-4d),
and plagioclase (Fig. 4e). The resulting compositions from these calculations are plotted
on the YSL ternary plots (Fig. 4b; Fig. 4d) and TKL ternary plots (Fig. 4a; Fig. 4c; Fig.
4e). The crystallization value of 10% for the mineral phase olivine is more representative
of the low degrees of fractional crystallization at ultraslow spreading ridge segments (Fig.
16
1b), while the higher fractional crystallization percentages and the other mineral phases
are more representative of faster spreading ridge segments (Fig. 1a) (Walker et al., 1979).
The olivine and plagioclase crystallization diagrams (Fig. 4a; Fig 4b; Fig. 4e)
show the original DMM composition remaining nearly unchanged in its relative trace
element concentrations even after 10%, 30%, and 50% fractional crystallization. The
different points representing different percentages of crystallization almost all nearly
overlap each other. In contrast, a more significant change is seen on the clinopyroxene
crystallization diagrams (Fig. 4c-4d), which shows that the points representing fractional
crystallization of a DMM composition to 10%, 30%, and 50% plot noticeably farther
away from the original DMM composition.
Importantly however, it should be noted that all of these mineral phases do not
simultaneously start crystallizing at the same time. Olivine is the first mineral to
crystallize from nearly all MORB magmas (Sobolev et al., 2007), so although the
crystallization of clinopyroxene (Fig. 4c-4d) and plagioclase (Fig. 4e) are also shown,
only the olivine crystallization diagrams (Fig. 4a-4b) are representative of the initial
degrees of fractional crystallization. The other mineral phases would not crystallize until
higher extents of crystal fractionation have occurred. At low degrees of fractional
crystallization, the number of mineral phase that crystallize may be minimized more than
at higher degrees of fractional crystallization.
Among the phases modeled in Figure 4, the next phases to crystallize after olivine
are plagioclase, followed by clinopyroxene. The crystallization of clinopyroxene happens
very late on the MORB liquidus (Walker et al., 1979) even at high pressures (Stolper,
1980). This means that clinopyroxene crystallization may affect the magma composition
17
(as seen on Figure 4) if the MORB magma is allowed to crystallize deeply at higher
pressure, where higher extents of fractional crystallization occurs (Elthon et al., 1992),
but the crystallization of clinopyroxene is unlikely to be a significant factor in affecting
MORB magma compositions in the low pressure, low degree fractionation that occurs at
ultraslow spreading ridges, such as Gakkel Ridge.
18
Figure 4a. Olivine crystallization diagram: YSL Ternary plot showing relative
concentrations of the trace elements Yb (ppm), Sm (ppm x 2) and La (ppm) of a depleted
MORB mantle composition (DMM) (Salters & Stracke, 2004), at 0%, 10%, 30%, and
50% fractional crystallization of the mineral phase olivine, respectively represented by a
black square, blue square, green square, and red square. The four squares almost overlap
each other. The partition coefficients of Yb, Sm, and La for olivine used in this
calculation were obtained from Salters & Stracke (2004) and are for 2 GPa pressure. The
global MORB array, comprising data from the MAR, EPR, SWIR, & SEIR, (PetDB;
Lehnert et al., 2000) is encircled in black.
19
Figure 4b. Olivine crystallization diagram: TKL Ternary plot showing relative
concentrations of the trace elements TiO2 (wt.% x 5), K2O (wt.% x 30) and La (ppm) of a
depleted MORB mantle composition (DMM) (Salters & Stracke, 2004) at 0%, 10%,
30%, and 50% fractional crystallization of the mineral phase olivine, respectively
represented by a black square, blue square, green square, and red square. The four
squares almost overlap each other. The partition coefficients of Ti, K, and La for olivine
used in this calculation were obtained from Salters & Stracke (2004) and are for 2 GPa
pressure. The global MORB array, comprising data from the MAR, EPR, SWIR, &
SEIR, (PetDB; Lehnert et al., 2000) is encircled in black.
20
Figure 4c. Clinopyroxene crystallization diagram: YSL Ternary plot showing relative
concentrations of the trace elements Yb (ppm), Sm (ppm x 2) and La (ppm) of a depleted
MORB mantle composition (DMM) (Salters & Stracke, 2004) at 0%, 10%, 30%, and
50% fractional crystallization of the mineral phase clinopyroxene, respectively
represented by a black square, blue square, green square, and red square. The partition
coefficients of Yb, Sm, and La for olivine used in this calculation were obtained from
Salters & Stracke (2004) and are for 2 GPa pressure. The global MORB array,
comprising data from the MAR, EPR, SWIR, & SEIR, (PetDB; Lehnert et al., 2000) is
encircled in black.
21
Figure 4d. Clinopyroxene crystallization diagram: TKL Ternary plot showing relative
concentrations of the trace elements TiO2 (Wt.% x 5), K2O (Wt.% x 30) and La (ppm) of
a depleted MORB mantle composition (DMM) (Salters & Stracke, 2004) at 0%, 10%,
30%, and 50% fractional crystallization of the mineral phase clinopyroxene, respectively
represented by a black square, blue square, green square, and red square. The partition
coefficients of Ti, K, and La for olivine used in this calculation were obtained from
Salters & Stracke (2004) and are for 2 GPa pressure. The global MORB array,
comprising data from the MAR, EPR, SWIR, & SEIR, (PetDB; Lehnert et al., 2000) is
encircled in black.
22
Figure 4e. Plagioclase crystallization diagram: YSL Ternary plot showing relative
concentrations of the trace elements Yb (ppm), Sm (ppm x 2) and La (ppm) of a depleted
MORB mantle composition (DMM) (Salters & Stracke, 2004) at 0%, 10%, 30%, and
50% fractional crystallization of the mineral phase plagioclase, respectively represented
by a black square, blue square, green square, and red square. The partition coefficients of
Yb, Sm, and La for olivine used in this calculation were obtained from White (2013). The
global MORB array, comprising data from the MAR, EPR, SWIR, & SEIR, (PetDB;
Lehnert et al., 2000) is encircled in black.
1.4.2) Partial melting
Upward convection of the mantle, which happens beneath the ridge axes at midocean ridges as a response to the spreading of the lithospheric plates, leads to the
23
decompression of the mantle, resulting in partial melting, forming MORB magmas (Fig.
1; Langmuir et al., 1992; Klein, 2003b). The partial melting process, which forms MORB
magmas, occurs within a melting region that extends from the base of the crust at ridge
axes down to as deep as 30 km to 100 km depth within the mantle (Hofmann, 1997).
Mid-ocean ridge partial melting does not melt mantle peridotite rocks entirely, but
instead only partially, with the proportion of mantle melting in the melting region ranging
as low as less than 1% (Zindler & Hart, 1986) to ranging between 10% and 20% (Snow et
al., 2001) or from 5% to 20% (Helffrich & Wood, 2001) or up to as high as 30% (Zindler
& Hart, 1986; Klein, 2003b). Since the proportion of melting is never as high as 100%,
this means that partial melting always leaves an unmelted solid refractory peridotite
residue remaining in the mantle after melt extraction (Hofmann, 1997; Klein, 2003b).
Partial melting is understood to have a less significant influence on the magmatic
composition than fractional crystallization (e.g., Klein 2003b), so in circumstances in
which fractional crystallization has a minimal effect (i.e., Fig 4a, Fig. 4b, Fig. 4d), partial
melting would also be expected to have a minimal effect. For instance, mid-ocean ridges
with an ultraslow spreading rate are settings where it has been suggested that the effects
of fractional crystallization and partial melting on magmatic compositions are both
minimized (Michael et al., 2003; Standish et al., 2008; Nauret et al., 2011).
1.4.3) Spreading rate effects
One parameter that has been suggested to control the degree to which partial
melting, fractional crystallization, and magmatic mixing occurs is the mid-ocean ridge
spreading rate, which is believed to be proportional to the extent of the melting column
24
(Michael et al., 2003; Standish et al., 2008; Nauret et al., 2011). Spreading at slower
spreading ridges has been suggested to involve more conductive cooling from the
surface, which produces thinner crust, thickens the lithosphere, and depresses the top of
the melting column (Fig. 1b; Michael et al., 2003; Standish et al., 2008; Nauret et al.,
2011). Consequently, the degrees to which magmatic differentiation processes occur
along ultraslow spreading ridges are lower compared to faster spreading ridges (Michael
et al., 2003; Standish et al., 2008; Nauret et al., 2011). For example, the degree of partial
melting drops sharply at spreading rates slower than 1.6 cm/year (Snow et al., 2001).
Also minimized are some processes that occur during melt migration, such as magma
mixing. Consequently, the geochemical signatures of mantle components in basalts from
ultraslow spreading ridges are less obscured by magmatic differentiation processes
(Michael et al., 2003).
Gakkel Ridge, located in the Arctic Basin, is an example of an ultraslow
spreading center, where the extents to which the processes of partial melting and
fractional crystallization operate are believed to be minimized (Michael et al., 2003).
Figure 4 has demonstrated that the low degrees of fractional crystallization – as expected
at ultraslow spreading ridges, such as Gakkel Ridge - has a minimal effect on the
magmatic composition, so it can thus be inferred that partial melting has an even more
negligible influence on MORB composition (Klein, 2003b). Thus, the effects of partial
melting do not need to be calculated for Gakkel Ridge, since it has already been
demonstrated that the more dominant process of fractional crystallization has little effect
on the varying degrees of enrichment in Gakkel Ridge MORBs.
25
1.5)
Mantle heterogeneity
Although magmatic differentiation processes can produce a range of differences
in basalt chemical compositions, differentiation of a single source component
composition alone cannot account for the full diversity of differences in chemical
composition observed in basalts, such as different amounts of enrichment in different
incompatible trace elements, because varying degrees of differentiation affecting a single
homogenous mantle source composition can only produce too a limited range of possible
basalt compositions – more limited than the wide degree of variation observable in global
basalt element compositions, as demonstrated by Figure 4.
A widely discussed view (e.g., Holness & Richter, 1989; Hofmann, 1997) is that
in order to explain the full variety of chemical compositions observed in basalts, it is
necessary to consider the presence of compositional heterogeneity in the mantle in the
form of multiple magmatic source components, each with distinct mineralogical and/or
chemical compositions, some of which are more enriched in certain elements than others.
These chemical heterogeneities may range in scale from as small as centimeters
(mineralogical scale) (Zindler & Hart, 1986; Sun & McDonough, 1989) or less than one
kilometer (Holness & Richter, 1989) or kilometers (Liu et al., 2008) up to as large as
thousands of kilometers (Zindler & Hart, 1986; Holness & Richter, 1989; Sun &
McDonough, 1989) or the size of ocean basins (Liu et al., 2008).
1.5.1) Recycling
One idea for the origin of the mantle’s heterogeneity involves the recycling of
crustal material, such as continental crust, oceanic crust, or sediments, into the mantle.
26
Recycling is the process by which crustal rocks and associated sediments are subducted
or delaminated and become incorporated into the mantle (Hofmann & White, 1982;
Holness & Richter, 1989; Helffrich & Wood, 2001). At the increased temperatures and
pressures at increased depths in the mantle, the subducted oceanic crust would transform
into eclogite, a denser material than the surrounding mantle, causing the subducted crust
to detach from the lithosphere and sink deep into the mantle (Hofmann & White, 1982).
Recycled components may become unstable and melt as a consequence of the increased
temperatures at increased depths in the mantle (Hofmann & White, 1982). The recycled
components may then be combined with the surrounding mantle by convective mixing
(Holness & Richter, 1989; Klein, 2003b). Recycled material adds chemical heterogeneity
to the mantle by introducing incompatible trace elements back into the mantle (Hofmann
& White, 1982). Incompatible elements are generally more concentrated in crustal
material, instead of in mantle rocks, because they were extracted from the mantle and
concentrated into the crust during ancient partial melting (Hofmann, 1988; Helffrich &
Wood, 2001; Hofmann, 1988; Hofmann, 2007).
Recycled material may be subducted to depths as deep as the core mantle
boundary and then brought back up by mantle convection (Hofmann & White, 1982; Van
Keken et al., 2002) or it may be material that has remained relatively shallow in the
mantle, such as subcontinental lithospheric mantle (SCLM) (Nauret et al., 2010). The
recycled material in the mantle is suggested to be dominantly composed of subducted
oceanic crustal and lithospheric material; whereas, the amount of recycled continental
material in the mantle is believed to be relatively smaller in comparison (Hoffman, 1997;
Hofmann, 2007).
27
1.5.2) Veins
One form in which enriched heterogeneous recycled components have been
suggested to exist in the mantle is in the form of “veins” of enriched material, composed
of more melt fertile and more trace element enriched rock-types, embedded within the
more abundant surrounding peridotite mantle (Wood, 1979; Dick et al., 2003; Nauret et
al., 2011).
Mantle veins may be composed of pyroxenite or eclogite, both of which are rocktypes that can form from subducted crustal material (Sobolev et al., 2005). Eclogite,
which is composed of clinopyroxene and garnet, is the rock type that subducted oceanic
basalts and gabbros transform into after being subducted (Hofmann & White, 1982;
Sobolev, 2005 et al.; Sobolev et al., 2007). Pyroxenite is a rock type that may be
produced when melted eclogite reacts with the peridotite (Sobolev et al., 2005; Sobolev
et al., 2007). Pyroxenite and eclogite both have a lower solidus than peridotite, so these
rock types start melting at higher pressure and greater depths than peridotite (Sobolev et
al., 2005; Sobolev et al., 2007). In fact, a pyroxenite composition can melt at 35 to 50 km
deeper than peridotite and can be 60% molten at the peridotite solidus (Dick et al., 2003).
This means that veins of eclogite and pyroxenite in the mantle would melt completely
during the MORB partial melting process, unlike the partial melting of peridotite, which
leaves behind a refractory unmelted residue in the mantle (Sobolev et al., 2005). The
100% melting experienced by eclogite and pyroxenite components may explain the
absence of veins composed of these rock types in abyssal peridotites recovered from the
seafloor.
28
The existence of mantle veins remains a topic of debate, but their presence in the
mantle would explain how sources of incompatible trace element enriched chemical
signatures can be contained in the mantle, even while the upper mantle is dominantly
composed of the more typical depleted MORB mantle source (Wood, 1979; Dick et al.,
2003; Nauret et al., 2011). The lower degrees of partial melting expected to be associated
with slower, means that an ultraslow spreading ridge with may sample melts from fertile
enriched veins preferentially over the depleted mantle, resulting in enriched basalts
(Snow et al., 2001; Liu et al., 2008).
1.5.3) Magmatic mixing
Mixing between magmatic melts derived from different mantle components can
affect MORB compositions (Le Roex et al., 1992; Nauret et al., 2011). Mixing between
magmatic melts is a process that operates as the melts derived from mantle partial
melting rise upwards from the mantle prior to erupting at the surface. Following
extraction from the mantle, these magmas mix together during their ascent, before
ultimately cooling to form basalts. If the different mantle source components have
distinct chemical compositions, the magmatic melts that they produce would also feature
differences in chemical composition. When these different melts mix together, the
chemical compositions of the basalts that ultimately form from them would end up
reflecting an intermediate composition between the different compositions of the mantle
source components from which they were derived. (Le Roex et al., 1992; Nauret et al.,
2011).
29
Mixing lines are clearly observable in basalt data plotted on the ternary plots, such
as Figure 3, which shows arrays of basalt composition data that plot between two
endmember points. Such data arrays can be interpreted to represent mixing lines, showing
the basalts resulting from the magmatic mixing that occurs between two endmember
compositions. By examining the endpoints of these ternary plot mixing lines, the
compositions of the endmember components in the mantle from which the magmatic
melts were derived can be interpreted.
1.5.4) OIB mantle components
Although the various heterogeneous components of the mantle are not well
defined and may have not all been identified, attempts have been made to identify certain
components, based on their isotopic signatures. A large amount of these studies have
focused on enriched ocean island basalts (OIB), because their enriched signatures are
more distinct than those of different enriched MORBs. Most of this previous work on
identifying mantle components has primarily focused on isotopes, not trace elements.
Three isotopically enriched mantle source components have been classified based
on the chemical compositions of enriched OIBs (Zindler & Hart, 1986). The three
defined OIB components along with DMM can together be considered to constitute four
major endmember components in the mantle, each of which produce basalts featuring
distinct isotopic signatures.
The three isotopically enriched OIB components are: enriched mantle 1 (EM1),
enriched mantle (EM2), and high μ (HIMU) (Zindler & Hart, 1986; Sun & McDonough,
1989). Although the EM1, EM2, and HIMU components were defined based on OIB
30
element concentration data instead of MORB, they are of interest to MORB studies,
because MORB trace element patterns may reveal information about their origins (Fig. 5;
Weaver, 1991a; Weaver, 1991b; Hofmann, 1997); thus, comparing MORB trace element
patterns to trace element patters of OIB components may yield information about the
origins of MORB source components.
The chemical signatures of EM1, EM2, and HIMU all feature relatively enriched
trace element concentrations compared to other oceanic basalts (Fig. 5; Hofmann, 1997).
Their chemical signatures also feature differences from each other, reflecting their
different origins (Weaver, 1991a; Weaver 1991b). Basalts derived from the EM1 and
HIMU components feature trace element patterns that are similar to each other and
resemble the trace element pattern of DMM-derived ocean crust MORB. Both of these
components predominantly reflect a recycled oceanic crust signature, with the EM1
component also showing contamination from pelagic sediment (Weaver, 1991a; Weaver
1991b).
In contrast, the EM2 component produces basalts with a distinctive trace element
pattern resembling not only ocean crust, but also continental crust (Hofmann, 1997). The
EM2 component is also dominated by a recycled oceanic crust signature, just as HIMU
and EM1 are, but the EM2 signature features the distinction of containing additional
contamination by continental terrigenous sediment (Weaver, 1991a; Weaver 1991b).
31
Figure 5a. NKL Ternary plot showing relative concentrations of Na 2O (wt.% x 5), K2O
(wt.% x 30), & La (ppm) in EM1, EM2, and HIMU basalt glass and whole rock samples.
OIB composition data are from GEOROC (Lehnert et al., 2000). EM1, EM2, and HIMU
datasets have been filtered to include only ICP-MS (inductively coupled plasma mass
spectrometer) trace element data. Also plotted are DMM (Salters & Stracke, 2004) and
PM (Sun & McDonough, 1989). Average N-MORB, T-MORB, and E-MORB are
calculated from basalt glass data from PetDB (Lehnert et al., 2000). The global MORB
array, comprising data from the MAR, EPR, SWIR, & SEIR, (PetDB; Lehnert et al.,
2000) is encircled in gray.
32
Figure 5b. TKL Ternary plot showing relative concentrations of TiO2 (wt.% x 5), K2O
(wt.% x 30), & La (ppm) in EM1, EM2, and HIMU basalt glass and whole rock samples.
EM1, EM2, and HIMU datasets have been filtered to include only ICP-MS trace element
data. The global MORB array is encircled in gray. Data are from the same references as
those in Figure 5a.
33
Figure 5c. YSL Ternary plot showing relative concentrations of Yb (ppm), Sm (ppm x
2), & La (ppm) in EM1, EM2, and HIMU in basalt glass and whole rock samples. EM1,
EM2, and HIMU datasets have been filtered to include only ICP-MS trace element data.
The global MORB array is encircled in gray. Data are from the same references as those
in Figure 5a.
2)
SETTING
Gakkel Ridge is 1800 km long. (Fig. 6; Fig. 7). It is located in the Arctic Basin
(Fig. 6). It is the northernmost segment of the global mid-ocean ridge system (Klein,
34
2003a). Gakkel Ridge is the active plate boundary across which the North American and
Eurasian Plates separate in the Arctic Basin (Fig. 6; Goldstein et al., 2008).
The western end of Gakkel Ridge is located at 82°N 2°W (Mühe et al., 1991;
Mühe et al., 1997) near Greenland, where Gakkel Ridge connects with the Lena Trough
(Fig. 6). The eastern end of Gakkel Ridge is located at 80°N 125°E (Mühe et al., 1991;
Mühe et al., 1997), or about 60 kim from the Laptev seashelf edge (Engen et al., 2003),
where the ridge terminates at the continental Siberian margin in the Laptev Sea (Fig. 6).
Figure 6. Bathymetric map showing the location of Gakkel Ridge and surrounding
features in the Arctic Basin, including Lena Trough (LT), Knipovich Ridge (KR), and
Lomonosov Ridge (LR) (Cochran, 2008). Spitsbergen island of the Svalbard archipelago
is located south of western Gakkel Ridge.
35
Gakkel Ridge is the world’s slowest spreading segment of the global mid-ocean
ridge system (Mühe et al., 1991; Klein, 2003a; Michael et al., 2003; Goldstein et al.,
2008; Liu et al., 2008; Shaw et al., 2010). The spreading rate is not consistent along its
length, but instead decreases from west to east along its axis. The full spreading rate
along Gakkel Ridge is the fastest at the western end of the ridge, where the full spreading
rate is 1.5 cm/year (Mühe et al., 1997) or 1.46 cm/year (Michael et al., 2003; Liu et al.,
2008) or ~1.4 cm/year (Shaw et al., 2010) or 1.3 cm/year (Jokat et al., 2003) or 1.28
cm/year (Cochran, 2008) or 1.27 cm/year (Dick, 2003). Spreading is slowest at the
eastern Siberian end of Gakkel Ridge, where the full spreading rate is as low as 0.8
cm/year (Mühe et al., 1997) or 0.65 cm/year (Cochran, 2008) or 0.63 cm/ year (Michael
et al., 2003; Liu et al., 2008) or 0.6 cm/year (Dick et al., 2003; Jokat et al., 2003) or ~0.6
cm/year (Shaw et al., 2010) or 0.3 cm/year (Klein, 2003a).
Gakkel Ridge is classified as an ultraslow spreading ridge (Fig. 1b). Ultraslow
spreading ridges are defined as having full spreading rates of less than 2 cm/year (Dick et
al., 2003; Jokat et al., 2003; Snow & Edmonds, 2007).
2.1)
Magmato-tectonic zones
Michael et al., 2003 divided Gakkel Ridge into three “magmato-tectonic”
segments, the Western Volcanic Zone (WVZ), the Sparsely Magmatic Zone (SMZ), and
the Eastern Volcanic Zone (EVZ), each with different magmatic behavior and differing
relative abundances of basalt and peridotite. Basalts recovered from each zone have
distinctly different chemical characteristics (Fig. 11; Fig. 12; Table 3; Table 4).
36
There is a segmentation of relative abundances of rock-types along Gakkel Ridge,
even though the it features no significant transform offsets (Goldstein et al., 2008) or
major fracture zones (Michael et al., 2003) to form mantle flow and compositional
boundaries along its length. This rules out physical boundaries as a possibility to explain
the chemical segmentation in the mantle along the ridge.
The variations in mantle source components in basaltic rocks recovered along
Gakkel Ridge do not vary in a consistent manner in correlation with the consistently
changing spreading rate along the ridge length (Klein, 2003a; Michael et al., 2003;
Goldstein et al., 2008). This rules out spreading rate variation as the primary factor
controlling magma compositions, and thus basalt compositions, along the ridge.
Because the segmentation of the ridge cannot be directly attributed to transform,
fractures, or spreading rate variation along the ridge, the segmentation instead must be
controlled by variations within the mantle (Michael et al., 2003; Goldstein et al., 2008)
Mantle characteristics that might be varying along the ridge may include different mantle
temperatures or different mantle compositions, such as compositions with different
melting fertility (Michael et al., 2003).
Each magmato-tectonic zone along Gakkel Ridge features multiple volcanic
centers (Fig. 7), but the abundances of volcanic centers and the distance of the spacing
between volcanic centers varies from zone to zone. Volcanic centers range in size from
15km to 50 km in diameter, typically 30 km along strike (Michael et al., 2003). The
spacing between volcanic centers ranges from 50km to 160 km apart (Michael et al.,
2003).
37
No basalt samples have been recovered east of the Eastern Volcanic Zone (EVZ)
(Michael et al., 2003). One reason for this is that the eastern end of Gakkel Ridge is
buried by sediment (Mühe et al., 1991; Engen et al., 2003). Specifically, the region
between the 69°E and 85°E volcanic centers and the region eastward beyond the 85°E
volcanic center is mostly covered in sediments (Cochran, 2008).
2.1.1) Western Volcanic Zone (WVZ) (7°W - 3°E)
The westernmost segment of the Gakkel Ridge is the Western Volcanic Zone
(WVZ). It is 220 km long (Cochran, 2008), extending from its western end at 83°N, 7°W,
where it connects with the northern Lena Trough, to 84°30’N, 3°E, where it transitions
into the sparsely magmatic zone (SMZ) (Fig. 7, Michael et al., 2003; Cochran, 2008).
The WVZ features numerous closely spaced volcanic centers and abundant basalt
on the seafloor (Michael et al., 2003; Cochran, 2008; Goldstein et al., 2008).
It should be noted that although the morphological distinction between Northern
Lena Trough (NLT) and westernmost Gakkel Ridge is clearly defined (Snow et al.,
2011), there is no clear geochemical boundary between the two; thus, the samples in this
thesis labeled as belonging to NLT may also be considered to belong to the western end
of the WVZ and vice versa.
2.1.2) Sparsely Magmatic Zone (SMZ) (3°E - 29°E)
The Sparsely Magmatic Zone (SMZ) is 275 km long (Michael et al., 2003;
Cochran, 2008), extending from its western end near 84°38’N, 3°E, where it connects
38
with the WVZ, to its eastern end at 86°N, 29°E, where it connects with the Eastern
Volcanic Zone (EVZ) (Fig. 7, Michael et al., 2003; Cochran, 2008).
Along most of its length, the SMZ is an amagmatic segment, with rock types on
the seafloor along its length consisting of mostly abyssal peridotite and few basalts
(Michael et al., 2003), which tend to be concentrated in along-ridge bathymetric highs.
Between 8°E and 12°E a few rare basalt samples were recovered, but these were diabases
or old fragments without glass, so their chemistry was not analyzed (Michael et al.,
2003). The only volcanic areas within the SMZ where basalt samples were recovered are
two separate volcanic centers located at 13°E and at 19°E. (Fig. 7; Michael et al., 2003;
Goldstein et al., 2008; Cochran, 2008). These volcanic centers are labeled in this thesis as
SMZ13 (13°E) and SMZ19 (19°E)
2.1.3) Eastern Volcanic Zone (EVZ) (29°E - 85°E)
The Eastern Volcanic Zone is at least 580 km long (Michael et al., 2003; Cochran,
2008). It extends from 29°E eastward (Fig. 7). Its eastern end is not well defined, either
ending at 85°E (Michael et al., 2003), or extending as far eastward as 96°E (Cochran et
al., 2008). Among the well identified volcanic centers of Gakkel Ridge, the most eastern
volcanic center is at 85°E (Fig. 7a; Fig. 8), so in Fig. 7a, the eastern end of the EVZ is
defined at 85°E, instead of 96°E.
The EVZ consists of widely spaced large volcanic centers connected by
amagmatic segments (Fig. 7; Michael et al., 2003). Michael et al. (2003) defined six areas
of concentrated volcanism within the EVZ centered at 31°E, 37°E, 43°E, 55°E, 69°E, and
85°E (Fig. 7a-7c). Additional volcanic centers within the EVZ have been suggested to
39
exist, including one at 92°-93°E (Dick et al., 2003; Cochran, 2008) and one in the
sediment covered region between the 69°E and 85°E volcanic centers (Cochran, 2008).
However these possible additional volcanic centers are not well defined, because no
dredging has been conducted between the 69°E and 85°E volcanic centers, and because
no basalts were recovered eastward beyond the 85° E volcanic center; thus, only the six
volcanic centers defined by Michael et al. (2003) will be considered in this study. These
six volcanic centers are labeled in this thesis from west to east as EVZ31 (31°E), EVZ37
(37°E), EVZ43 (43°E), EVZ55 (55°E), EVZ69 (69°E), and EVZ85 (85°E).
2.2)
Adjacent Arctic Basin features
At its western end, the Gakkel Ridge is connected to the northern end of Lena
Trough (Fig. 6; Snow et al., 2011). Lena Trough is divided into two sections with
chemically distinct basalts, Northern Lena Trough (NLT) and Central Lena Trough
(CLT) (Nauret et al., 2011). Like Gakkel Ridge, Lena Trough is an ultraslow spreading
ocean ridge. The southern end of Lena Trough connects with Knipovich Ridge and from
there, the ridge system extends beyond the Arctic Basin and connects with the rest of the
Mid-Atlantic Ridge system (Fig. 6) (Engen et al., 2003). Spitsbergen is an island located
to the south of the WVZ (Fig. 6) and is part of the Svalbard archipelago.
3)
BACKGROUND
The earliest analyses of Gakkel Ridge MORB samples were conducted on a small
number of whole rock samples (Mühe et al., 1991; Mühe et al., 1993; Mühe et al., 1997)
recovered from early Arctic expeditions. However, the data in these studies included only
40
basalt glass samples; thus, data obtained from analyses of whole rock samples, such as
those of the earliest Gakkel Ridge MORB analyses, are not examined in this study.
The Gakkel Ridge data examined in this study are from rock samples recovered,
mostly by dredging, from later expeditions, one of which was the Arctic Mid Ocean
Ridge Expedition (AMORE) 2001 cruises by the research icebreakers, US Coast Guard
Cutter Healy and PFS Polarstern of the Alfred Wegener Institute for Polar and Marine
Research of Bremerhaven, Germany (Michael et al., 2003; Snow & Edmonds, 2007)
Later, additional samples were collected during the Arctic Gakkel Vents Expedition
(AGAVE) 2007 cruise, conducted by the Woods Hole Oceanographic Institution, on the
Swedish icebreaker Oden (Shaw et al., 2010).
The dredging stations where samples were collected during the AMORE 2001
cruises are shown on Figure 7. The dredging locations of samples collected during the
AGAVE 2007 cruise are shown on Figure 8. No samples were collected farther east than
the EVZ85 volcanic center (Fig. 7; Fig. 8).
3.1)
AMORE 2001
Basalt samples collected by AMORE 2001 span across a range in longitude from
8° W to 86° E across a distance of 850 km along the ridge at over 200 dredging stations
(Fig. 7; Michael et al., 2003; Snow et al., 2002). Goldstein et al. (2008) analyzed the
chemistry of these samples and reported major element, trace element and isotope
composition data of glasses of 38 samples: 16 WVZ samples, 3 SMZ13 samples, 6
SMZ19 samples, 2 EVZ31 samples, 2 EVZ37 samples, 2 EVZ43 samples, 3 EVZ69
samples, and 2 EVZ85 samples.
41
Michael et al. (2003) reported data for an additional 1 WVZ sample, 1 SMZ13
sample, and 1 EVZ85 sample not included in the data set of Goldstein et al. (2008).
The published data reported by Michael et al. (2003) and Goldstein et al. (2008)
include only about one third of the total number of samples collected by AMORE 2001.
This study examines 68 additional AMORE 2001 samples, ranging from SMZ13 to
EVZ55, that were collected, but had not yet been analyzed until now (Table 3, Table 4).
Figure 7 is a bathymetric map showing the Gakkel Ridge as studied by the
AMORE 2001 expedition. Dredging stations where rock samples were collected are
represented as colored dots. The areas covered by each volcanic center along the ridge
are designated by boxes. Displayed above each of those boxes are the sample numbers of
basalt samples collected from each volcanic center.
42
43
Figure 7a. Gakkel Ridge AMORE 2001 sample location map (Snow et al., 2002).
44
Figure 7b. Gakkel Ridge AMORE 2001 sample location map (Snow et al., 2002).
45
Figure 7c. Gakkel Ridge AMORE 2001 sample location map (Snow et al., 2002).
46
Figure 7d. Gakkel Ridge AMORE 2001 sample location map (Snow et al., 2002).
47
Figure 7e. Gakkel Ridge AMORE 2001 sample location map (Snow et al., 2002).
3.2)
AGAVE 2007
All of the basalt samples collected and analyzed by AGAVE 2007 are from the
EVZ85 volcanic center. Shaw et al. (2010) analyzed and reported trace element
concentration data of 9 EVZ85 basalt glass samples.
Figure 8. Locations of samples collected during the AGAVE 2007 cruise at the EVZ85
volcanic center (Shaw et al., 2010).
48
3.3)
Earlier studies
3.3.1) Geochemistry of Gakkel Ridge MORBs
The Gakkel Ridge MORB geochemical data included on the plots in this study
include data from earlier studies by Michael et al. (2003), Goldstein et al., (2008), Shaw
et al., (2010), and Snow (in prep). Michael et al. (2003) and Goldstein et al. (2008) found
that even though it might be expected for enrichment levels to increase with decreasing
spreading rate, the isotope and trace element enrichment levels of MORBs along Gakkel
Ridge actually do not correlate with the spreading rate, which slows down from west to
east along the ridge. Instead, the enrichment levels in the Gakkel Ridge MORBs vary
inconsistently with the distance along the ridge axis.
Goldstein et al. (2008) identified an isotopic boundary in the middle of Gakkel
Ridge, apparently located in the middle of the SMZ, between the SMZ13 and SMZ19
volcanic centers. The boundary divides Gakkel Ridge into two isotopic provinces. To
illustrate this boundary, they showed that the isotopic signature of MORBs from west of
the isotopic boundary (i.e., WVZ, SMZ13) plots along a separate trend on a 143Nd/144Nd
vs. 87Sr/86Sr isotope ratio plot than MORBs from east of the isotopic boundary (i.e.,
SMZ19, EVZ).
One of the characteristics displayed by the geochemistry of MORBs of the
western province (i.e., WVZ, SMZ13), identified by Goldstein et al. (2008), is that they
feature a “Dupal anomaly” signature, which is typically characteristic of southern
hemisphere MORBs, not northern hemisphere MORBs. This makes MORBs from the
western part of Gakkel Ridge the only example of MORBs with a significant Dupal
anomaly signature in the northern hemisphere (Goldstein et al., 2008).
49
A Dupal anomaly is characterized by elevated Δ8/4 ratios (Hart, 1984). Δ8/4 is a
measure of the amount of deviation in the 208Pb/204Pb ratio from the Northern
Hemisphere Reference Line (NHRL), which is defined as a 208Pb/204Pb ratio
representative of MORBs of the northern hemisphere (Hart, 1984).
Goldstein et al. (2008) found that the MORBs of the western province (i.e., WVZ,
SMZ13) all feature a Δ8/4 ratio above 30, in contrast to Δ8/4 ratios of samples from the
eastern province, which are all below 30 (Fig. 10b). Goldstein et al. (2008) interpreted
that SCLM from Spitsbergen is the source of the Dupal anomaly signature in western
Gakkel Ridge. In contrast, the eastern isotopic province of Gakkel Ridge, with its lower
Δ8/4 ratio, does not feature a Dupal anomaly signature and instead resembles north
Atlantic MORB; thus, does not contain SCLM in its source (Goldstein et al., 2008; Liu et
al., 2008).
Shaw et al., (2010) analyzed the trace elements compositions in basalt glass
samples from the EVZ85, but they did not measure their major element compositions.
Consequently, EVZ85 data are capable of being plotted on the YSL ternary plots (e.g.,
Fig. 12c) but not on the NKL and TKL ternary plots (e.g., Fig. 12a-12b), because no
Na2O, K2O, and TiO2 concentration data are currently available for these samples.
3.3.2) Geochemistry of other Arctic Basin basalts
Basalts of the Lena Trough, particularly the CLT, possess unique isotopic (Fig.
10) and trace element (Fig. 11) characteristics, primarily characterized by being more
alkalic than any other basalts ever dredged from the ocean floor (Snow et al., 2001). This
trace element signature indicates the presence of a unique K-enriched CLT mantle source
50
component, likely to be SCLM in origin (Nauret et al., 2011). This compositionally
distinct CLT component appears to be spatially widespread in the Arctic Basin mantle,
being present along Gakkel Ridge as well, not simply confined to the CLT area, because
the NLT and WVZ basalts also appear to be influenced by the CLT component (Fig. 12;
Snow J.E., in prep).
The Spitsbergen basalt composition appears to constitute an enriched endmember
composition that forms a trend with the WVZ compositions on isotope plots (Fig. 10;
Goldstein et al., 2008). The enrichment of the Spitsbergen composition possibly indicates
the presence of a SCLM component in the Arctic mantle (Goldstein et al., 2008; Nauret
et al., 2011).
Even though the Knipovich Ridge (i.e., the mid-ocean ridge that connects to the
southern end of Lena Trough) is geographically relatively close to Lena Trough, isotope
ratios of MORBs from Knipovich Ridge, which were compiled from PetDB (Lehnert et
al., 2000), when plotted on the 143Nd/144Nd vs. 87Sr/86Sr plot (Fig. 10a), aligns with the
SMZ19-EVZ array in eastern Gakkel Ridge, not the western WVZ-SMZ13 array. The
Δ8/4 values of Knipovich MORBs are also within range of the eastern isotopic province,
not the western province.
51
A
B
Isotopic boundary
Figure 9. Radiogenic isotope ratios of Arctic Basin basalt glasses. A) 143Nd/144Nd vs.
87
Sr/86Sr of Arctic basalt glasses. B) Δ8/4 of Gakkel Ridge samples, with basalts from
other locations plotted at 0 km distance for comparison. Δ8/4 measures the amount of
change in the 208Pb/204Pb ratio from a northern hemisphere reference line, which is an
52
average 208Pb/204Pb ratio in northern hemisphere basalts (Hart, 1984). Plotted samples are
basalts from Gakkel Ridge (Michael et al., 2003; Goldstein et al., 2008). Samples plotted
at “zero distance” are included for comparison – these samples are from Lena Trough
(Nauret et al., 2011), Knipovich (PetDB; Lehnert et al., 2000), Spitsbergen (Ionov et al.,
2002), and Macquarie Island (Kamenetsky et al., 2000), the last of which are included as
a representative E-MORB composition. The isotopic boundary in the middle of the SMZ,
described by Goldstein et al. (2008), is represented by a vertical red dashed line.
(Modified from Goldstein et al., 2008).
4)
METHODS
Major element composition analyses were performed by electron microprobe at
the University of Houston and at the JEOL 7900 electron probe facility of the MaxPlanck Institute for Chemistry in Mainz, Germany. Trace element composition analyses
were performed by laser ablation inductively coupled plasma mass spectrometry (LAICP-MS) at the University of Houston.
Analyses were performed on mounted basalt glass chips. The major and trace
element composition data reported in this study were averaged from three to four spots on
two separate glass chips per sample to ensure reproducibility of the measured data. Major
and trace elements were measured on the same glass chips.
All Gakkel Ridge data reported in this study were measured from basalt glass
samples, not whole rock samples. The other data included on the diagrams of this study
are mostly glass data, but some whole rock data were included from locations from which
glass data were unavailable. The accompanying figure captions point out which data are
whole rock data. Glass data are preferred over data measured from whole rock analyses
to help minimize possible contamination of in whole rock compositions, such as by
seawater. However, Brodholt and Batiza (1989) pointed out in their analysis of global
53
MORB data that plotting glass and whole rock analyses separately showed no difference
between the two subsets. Thus, although glass data were plotted whenever possible as an
extra precaution for this study, it may not have actually made a significant difference.
Major and trace elements from a total of 68 Gakkel Ridge samples were analyzed
for this study: 27 SMZ13 samples, 17 SMZ19 samples, 5 EVZ31 samples, 7 EVZ37
samples, 2 EVZ43 samples, and 10 EVZ55 samples. The figures in this thesis combine
data from these newly analyzed Gakkel Ridge basalt samples with all the previously
analyzed Gakkel Ridge basalt glass samples (Michael et al., 2003; Goldstein et al., 2008;
Shaw et al., 2010).
4.1)
LA-ICP-MS
Laser ablation inductively coupled plasma mass spectrometry is an analytical
procedure for determining the trace element concentrations of solid samples. The laser
ablates particles from spots on the surface of the glass of the basalt samples. These
ablated particles are then ionized by a plasma torch, producing charged ions that are then
separated according to their mass by a quadrupole mass spectrometer, allowing for
concentration data for each trace element to be recorded.
The LA-ICP-MS system used for alayzing trace elements, along with the major
element Ti, comprised of a CETAC LSX-213 Laser Ablation System coupled with a
Varian ICP-MS. Analyses were performed using a 50 μm beam diameter spot size. The
quality of the data was checked and ensured by repeated measurements of the basalt glass
standards KL2-G and BIR-1 (Appx. 6). Standard compositions were obtained from the
54
Geological and Environmental Reference Materials (GeoReM; Jochum et al., 2005). One
measurement of a standard was made for every two sample measurements.
4.2)
Electron microprobe
An electron microprobe works by bombarding the polished surfaces of the basalt
glass samples coated with a thin layer of carbon with a beam of high-energy electrons.
The electron beam ionizes atoms within several microns of the sample surface, removing
electrons from inner orbitals of the atoms, producing electron orbital vacancies, causing
electrons from higher energy orbitals to drop down into the lower energy vacancies; thus,
emitting photons of characteristic energies and wavelengths for each element; thus,
allowing element concentrations to be measured after counts are corrected for atomic
number, absorption, and fluorescence effects.
The electron microprobe used for analyzing the major elements of the Gakkel
Ridge basalt samples was a CAMECA SX50. Major element analyses were performed
using a 15 kV accelerating voltage, ~2 micron spot size, and 20 nA beam intensity. The
secondary standard used was VG-2 (Appx. 6; Jarosewich, 1980; Jarosewich, 2002).
4.3)
Uncertainty
Precision is a measure of the reproducibility of a measurement. Repeated
measurements that produce similar values are more precise; repeated measurements that
deviate from each other to a greater extent are less precise. Analyses with the greatest
possible level of precision would reproduce the exact same value every time the same
measurement is repeated. The amount of deviation from one measurement to another is
55
defined as the amount of uncertainty or error. Analyses are considered to be of better
quality when uncertainty is minimized.
Accuracy is defined as the closeness of a measured value to the true or accepted
value. Accuracy was constrained by analyzing basalt standards as unknowns during the
electron microprobe and LA-ICP-MS analyses. These check standard measurements were
compared to the published accepted values of those standards. The measured elements
were found to be within the analytical error of the true values (Appx. 6).
The uncertainty of the analyses is represented on the figures of this thesis as
“uncertainty polygons” (i.e., uncertainty fields) surrounding each plotted data point. The
extents covered by these polygons show the degree of precision for each data point
measurement that is plotted on the ternary plots. Uncertainty polygons were calculated by
first calculating the standard deviations of the analyses of the independent check
standards for all three elements of each ternary plot. Those calculated standard deviations
are then added and subtracted to and from the average concentrations of each element in
each sample, in order to calculate the points of the uncertainty polygons surrounding each
plotted sample.
56
A
std
199-1-2
+++
+++-+
-++
--+
-++----
SMZ19: 199-1-2 Uncertainty
139
La
K2O
TiO2
0.01
0.02
0.00
0.08
2.17
1.21
0.09
2.20
1.21
0.09
2.20
1.20
0.09
2.15
1.21
0.07
2.20
1.21
0.07
2.15
1.21
0.07
2.20
1.20
0.09
2.15
1.20
0.07
2.15
1.20
B
std
252-1
+++
+++-+
-++
--+
-++----
SMZ19: 252-1 Uncertainty
139
La
K2O
TiO2
0.01
0.02
0.61
9.80
0.62
9.82
0.62
9.82
0.62
9.77
0.60
9.82
0.60
9.77
0.60
9.82
0.62
9.77
0.60
9.77
0.00
1.39
1.39
1.39
1.39
1.39
1.39
1.39
1.39
1.39
Table 2. Average concentrations in ppm (yellow row) (averaged from four
measurements) of K2O, La, and TiO2 for the samples 199-1-2 (Table 2a) and 252-1
(Table 2b) from SMZ19, along with the corresponding standard deviations (std) (green
row), and the calculated results from adding and subtracting the standard deviations to the
average concentrations for each of the three elements (orange rows).
Table 2 shows the results from adding and subtracting the standard deviations of
the check standard analyses to and from the concentrations of the elements Sm, La, and
Yb for the samples 199-1-2 and 252-1 from the volcanic center EVZ55 to produce the
“uncertainty points” shown on Table 2 (orange rows). These points, when plotted on a
ternary plot (Fig. 10; Fig. 12), form a polygon around each data point. The field that is
covered by each polygon represents the field of uncertainty corresponding to each data
point. Two of the points (the “+++” and “---“ points) plot within the uncertainty polygon;
thus, do not define the polygon, so they have been removed from the ternary plots
showing the uncertainty polygons (Fig. 10; Fig. 12), leaving the polygon with six points,
forming six sides.
Figure 10 plots only two samples and for an unobstructed look at their
accompanying uncertainty polygons. The uncertainty polygons are defined by the black
dots surrounding the samples 199-1-2 and 252-1 from the volcanic center SMZ19,
57
produced from the uncertainty points calculated on Table 2. The uncertainty points of
sample 199-1-2 are more spread out and its resulting uncertainty polygon is larger than
the uncertainty points and uncertainty polygon of sample 252-1. This is a consequence of
sample 199-1-2 having smaller elemental abundances that sample 252-1. Smaller
elemental abundances yield lower accuracy than higher elemental abundance. As a
general trend, the samples plotting towards the depleted end of the global MORB array
(e.g., 199-1-2) have larger uncertainty polygons than samples plotting towards the
enriched end (e.g., 252-1).
Uncertainty polygons were calculated for all of the samples that were analyzed in
this study (Fig. 12; Fig. 13).
58
Figure 10. TKL ternary plot, featuring the elements La (ppm), Sm (ppm x 2), and Yb
(ppm), showing six uncertainty points, forming a six-sided polygon, surrounding each of
the plotted samples, 199-1-2 and 252-1 from SMZ19. Average MAR N-MORB and EMORB and global MORB array (circled in gray) data compiled from PetDB (Lehnert et
al., 2000).
59
4.4)
Two-component mixing lines
Two-component mixing lines were calculated using the equation:
XM=XAfA+XB(1-fA)
where XM denotes the concentration of element X in a two-component mixture, XA
denotes the concentration of X in component A, X B denotes the concentration of X in
component B, and fA denotes the abundance of component A in the given mixture. The
two-component mixing equation was obtained from White (2013). Mixing lines were
plotted on trace element ternary plots (Fig. Z).
5)
RESULTS
Results from this study are presented as element concentration tables (Table 3,
Table 4), a trace element spidergram (Fig. 11), and trace element concentration ternary
plots (Fig. 12).
Figure 11 shows averaged Gakkel Ridge basalt trace elements on a spidergram,
along with other averaged MORBs for comparison, including Lena Trough MORBs and
other enriched MORBs.
Gakkel Ridge basalts are all more enriched than the average MAR N-MORB.
This relative enrichment is visible on the trace element spidergram (Fig. 11), which
shows the averaged compositions from each volcanic center of Gakkel Ridge all plotting
higher than MAR N-MORB in nearly all the different trace elements. The average
enrichment in MORB is not consistent across Gakkel Ridge, but instead, varies from
60
volcanic center to volcanic center (Fig. 11). The volcanic centers SMZ13 and SMZ19 are
noticeably more enriched than the others along Gakkel Ridge (Fig. 11). Overall, however,
the level of enrichment in Gakkel Ridge MORBs does not appear to reach that of the EMORBs, such as average MAR E-MORB, Macquarie Island basalts, and SWIR-oblique
segment MORBs.
Figure 11. Spidergram showing relative trace element concentrations of Arctic basalt
glasses from the Central and Northern Lena Trough (CLT, NLT) (Nauret et al., 2011),
Gakkel Ridge (WVZ) (Snow J.E., in prep), SMZ, EVZ (Michael et al., 2003; Goldstein et
al., 2008), Southwest Indian Ridge oblique segment (SWIR-Obl) (Standish et al., 2008),
Macquarie Island (Kamenetsky et al., 2000), and average Mid-Atlantic Ridge E-MORB,
T-MORB, and N-MORB (Klein, 2003b). Elements are normalized to pyrolite
(McDonough & Sun, 1995). Elements are arranged in order of decreasing incompatibility
from left to right. (Modified from Shaw et al., 2010).
61
5.1)
Data
Major element concentrations of Gakkel Ridge basalt glass samples from the
volcanic centers SMZ 13, SMZ 19, EVZ 31, EVZ 37, EVZ 43, and EVZ 55 are shown in
Table 3. Trace element concentration data for those volcanic centers are shown in Table
4. The EVZ69 and EVZ85 volcanic centers were not a focus of this study, because
samples recovered from the region of Gakkel Ridge located east of the EVZ55 were not
analyzed for this study. All EVZ69 and EVZ85 basalt composition data included in the
figures in this study are from earlier studies (Michael et al., 2003; Goldstein et al., 2008;
Shaw et al., 2010). The chemistry of the WVZ and NLT is analyzed and discussed in
detail in another study (Snow J. E., in prep), so the WVZ and NLT are not a focus of this
study. WVZ and NLT basalt compositions that are included in the figures of this study
are from Snow J.E. (in prep).
62
5.1.1) Gakkel Ridge major element compositions
________________________________________________________________________
Sample
D35-69
D35-38
D35-4
D35-2
D35-34
244-10
244-2
244-SG
D37-8
D36-53
D36-63
D36-66
D36-64
D36-70
D36-68
D36-59
D36-62
D36-2
D36-45
D38-26
D38-25
D38-27
D38-29R
D38-12
D38-4
D38-29
248-59
199-1-1
199-1-2
199-8
D41-18
D41-17
D42-16
D42-18
D42-12
252-1
253-1
254-1
255-1
255-3
255-5
315-1
D45-10
260-1
263-12
263-19
263-16
D47-7
309-5
265-2A
265-2
265-5
307-8
266-15
266-13
304-3
304-2
297-2
297-36
297-1
297-SG
271-1-11
271-1-9
270-2
270-19
295-2
294-2
Zone
Lon
Lat
Distance Mg#
SMZ 13
10.71 85.27
308
0.62
SMZ 13
10.71 85.27
308
0.62
SMZ 13
10.71 85.27
308
0.62
SMZ 13
10.71 85.27
308
0.61
SMZ 13
10.71 85.27
308
0.62
SMZ 13
11.04 85.05
312
0.63
SMZ 13
11.04 85.05
312
0.58
SMZ 13
11.04 85.05
312
0.58
SMZ 13
11.39 85.30
316
0.56
SMZ 13
12.48 85.27
329
0.61
SMZ 13
12.48 85.27
329
0.61
SMZ 13
12.48 85.27
329
0.61
SMZ 13
12.48 85.27
329
0.61
SMZ 13
12.48 85.27
329
0.61
SMZ 13
12.48 85.27
329
0.61
SMZ 13
12.48 85.27
329
0.61
SMZ 13
12.48 85.27
329
0.61
SMZ 13
12.48 85.27
329
0.61
SMZ 13
12.48 85.27
329
0.61
SMZ 13
12.68 85.31
332
0.62
SMZ 13
12.68 85.31
332
0.62
SMZ 13
12.68 85.31
332
0.64
SMZ 13
12.68 85.31
332
0.64
SMZ 13
12.68 85.31
332
0.63
SMZ 13
12.68 85.31
332
0.64
SMZ 13
12.68 85.31
332
0.64
SMZ 13
13.45 85.21
341
0.66
SMZ 19
16.21 85.56
371
0.60
SMZ 19
16.21 85.56
371
0.61
SMZ 19
16.21 85.56
371
0.65
SMZ 19
17.95 85.69
390
0.67
SMZ 19
17.95 85.69
390
0.68
SMZ 19
18.09 85.73
391
0.64
SMZ 19
18.09 85.73
391
0.64
SMZ 19
18.09 85.73
391
0.63
SMZ 19
18.22 85.62
392
0.66
SMZ 19
18.30 85.71
393
0.63
SMZ 19
19.50 85.74
405
0.64
SMZ 19
20.28 85.79
413
0.66
SMZ 19
20.28 85.79
413
0.67
SMZ 19
20.28 85.79
413
0.67
SMZ 19
20.45 85.81
415
0.62
SMZ 19
25.14 86.03
459
0.63
EVZ 31
29.14 85.97
494
0.48
EVZ 31
30.66 86.06
507
0.59
EVZ 31
30.66 86.06
507
0.58
EVZ 31
30.66 86.06
507
0.64
EVZ 31
31.27 86.05
512
0.59
EVZ 37
36.01 86.23
551
0.60
EVZ 37
37.75 86.33
565
0.56
EVZ 37
37.75 86.33
565
0.57
EVZ 37
37.75 86.33
565
0.66
EVZ 37
38.16 86.36
568
0.64
EVZ 37
38.76 86.31
573
0.58
EVZ 37
38.76 86.31
573
0.64
EVZ 43
40.99 86.48
591
0.64
EVZ 43
40.99 86.48
591
0.64
EVZ 55
47.11 86.68
639
0.64
EVZ 55
47.11 86.68
639
0.65
EVZ 55
47.11 86.68
639
0.65
EVZ 55
47.11 86.68
639
0.65
EVZ 55
47.72 86.84
644
0.66
EVZ 55
47.72 86.84
644
0.66
EVZ 55
50.12 86.75
663
0.67
EVZ 55
50.12 86.75
663
0.68
EVZ 55
52.30 86.80
681
0.66
EVZ 55
56.06 86.89
713
0.65
MgO
6.66
7.36
7.36
7.36
7.38
6.58
6.65
6.76
6.27
6.45
6.46
6.48
6.48
6.49
6.53
6.53
6.55
6.59
6.62
7.00
7.00
7.39
7.39
7.43
7.46
7.49
7.60
8.17
8.19
9.29
7.64
7.86
7.49
7.50
7.52
7.25
6.89
6.98
7.38
7.40
7.41
6.96
8.63
5.55
7.06
7.29
8.16
6.76
7.31
6.65
6.78
8.20
7.48
6.84
7.92
7.81
7.82
8.74
8.81
8.98
9.08
7.90
8.26
8.41
8.60
7.80
8.19
SiO2
52.0
51.0
50.9
51.0
50.9
50.8
50.8
51.6
51.3
50.9
51.0
51.6
51.2
50.7
51.6
51.4
51.0
52.0
52.0
51.1
51.2
50.9
51.1
51.0
50.9
50.8
51.0
48.0
47.9
48.5
50.9
51.0
50.8
50.9
50.8
51.3
51.5
51.2
51.4
51.4
51.4
51.5
49.2
50.8
50.6
50.5
50.4
50.9
50.7
51.0
51.6
51.0
50.7
50.7
50.5
50.8
50.7
48.3
48.5
48.7
48.7
49.9
49.8
49.7
49.7
49.9
49.0
Al 2O3
16.3
15.6
15.6
15.6
15.6
16.5
15.8
16.0
15.7
16.6
16.6
16.5
16.6
16.5
16.5
16.5
16.5
16.2
16.2
16.5
16.5
16.5
16.6
16.5
16.6
16.5
16.6
17.4
17.6
16.6
16.4
16.7
16.6
16.6
16.6
16.9
16.0
16.1
16.5
16.5
16.6
16.1
16.2
14.3
15.8
15.4
16.6
15.8
15.8
15.4
15.5
16.5
16.5
15.6
16.0
16.2
16.2
17.9
17.7
17.7
17.6
16.9
16.4
16.8
16.7
16.9
17.2
Na 2O
3.77
3.63
3.70
3.74
3.58
3.36
3.56
3.49
3.91
3.83
3.73
3.95
3.84
3.92
3.83
3.74
3.84
3.83
3.65
3.64
3.63
3.52
3.57
3.54
3.46
3.49
3.41
2.82
3.01
3.09
3.34
3.27
3.42
3.48
3.51
3.68
3.57
3.37
3.37
3.37
3.43
3.51
3.29
3.79
3.38
3.56
3.02
3.62
3.45
3.63
3.27
3.23
3.25
3.57
3.42
3.33
3.34
2.84
2.88
2.74
2.88
3.39
3.52
3.31
3.42
3.34
3.41
K2O TiO2
0.659
1.53
0.469
1.64
0.478
1.64
0.483
1.64
0.479
1.65
0.744
1.71
0.292
1.63
0.342
1.63
0.463
1.78
0.655
1.53
0.653
1.51
0.652
1.51
0.657
1.52
0.646
1.54
0.648
1.52
0.643
1.54
0.645
1.52
0.651
1.52
0.647
1.51
0.326
1.50
0.328
1.50
0.280
1.43
0.284
1.40
0.283
1.41
0.280
1.40
0.290
1.39
0.337
1.27
0.055
1.16
0.080
1.21
0.277
1.19
0.240
1.17
0.246
1.14
0.363
1.34
0.370
1.34
0.372
1.34
0.611
1.40
0.370
1.44
0.401
1.41
0.340
1.22
0.346
1.22
0.336
1.23
0.324
1.50
0.167
1.44
0.424
2.33
0.232
1.77
0.327
2.04
0.283
1.35
0.357
1.76
0.353
1.79
0.242
1.78
0.254
1.80
0.241
1.46
0.337
1.60
0.395
2.06
0.383
1.72
0.247
1.56
0.241
1.55
0.134
1.06
0.174
1.14
0.150
1.05
0.144
1.06
0.342
1.49
0.252
1.47
0.390
1.55
0.389
1.55
0.333
1.45
0.113
1.38
CaO
9.81
9.74
9.77
9.77
9.82
10.4
9.84
9.73
9.55
9.83
9.89
9.93
9.87
9.93
9.90
9.87
9.85
9.86
9.84
10.6
10.6
10.6
10.7
10.7
10.7
10.7
10.4
11.0
10.9
10.1
11.3
11.5
10.6
10.7
10.5
10.1
11.0
11.3
11.3
11.3
11.4
10.8
10.4
9.71
10.3
10.2
11.1
10.9
10.3
10.6
10.8
10.9
10.8
10.1
10.2
10.7
10.8
11.0
10.9
11.0
10.9
10.8
10.9
10.8
10.7
11.0
11.0
FeO
8.24
9.08
9.06
9.14
9.11
7.81
9.53
9.59
9.78
8.22
8.23
8.18
8.27
8.19
8.18
8.20
8.19
8.19
8.22
8.56
8.54
8.40
8.35
8.49
8.44
8.41
7.84
10.7
10.5
9.87
7.48
7.43
8.43
8.53
8.57
7.47
8.11
7.77
7.41
7.38
7.32
8.43
10.1
11.8
9.86
10.5
8.91
9.22
9.62
10.3
10.3
8.28
8.39
9.82
8.92
8.71
8.71
9.65
9.36
9.58
9.57
8.09
8.52
8.16
8.16
8.08
8.73
MnO
0.15
0.16
0.17
0.16
0.17
0.15
0.18
0.18
0.18
0.16
0.15
0.16
0.14
0.14
0.15
0.15
0.15
0.16
0.15
0.17
0.18
0.17
0.17
0.14
0.16
0.17
0.15
0.17
0.19
0.15
0.14
0.15
0.15
0.16
0.16
0.15
0.17
0.14
0.14
0.13
0.13
0.17
0.18
0.22
0.18
0.19
0.16
0.18
0.18
0.19
0.17
0.15
0.15
0.19
0.17
0.17
0.16
0.18
0.16
0.17
0.17
0.15
0.17
0.15
0.15
0.15
0.17
Cl
0.042
0.037
0.037
0.035
0.036
0.078
0.018
0.018
0.029
0.044
0.046
0.042
0.043
0.044
0.044
0.042
0.045
0.045
0.046
0.017
0.018
0.015
0.015
0.018
0.019
0.016
0.022
0.009
0.008
0.013
0.010
0.010
0.022
0.021
0.022
0.038
0.018
0.022
0.020
0.018
0.015
0.016
0.010
0.014
0.010
0.012
0.013
0.013
0.015
0.011
0.014
0.009
0.014
0.021
0.017
0.008
0.011
0.004
0.006
0.005
0.007
0.012
0.010
0.013
0.012
0.011
0.006
S
0.091
0.097
0.095
0.089
0.101
0.083
0.106
0.100
0.113
0.080
0.083
0.091
0.082
0.084
0.087
0.086
0.085
0.072
0.084
0.104
0.096
0.097
0.096
0.100
0.099
0.096
0.070
0.096
0.099
0.100
0.090
0.078
0.084
0.096
0.083
0.082
0.094
0.089
0.085
0.084
0.080
0.101
0.116
0.145
0.105
0.111
0.095
0.099
0.107
0.131
0.119
0.087
0.097
0.117
0.095
0.088
0.104
0.093
0.089
0.090
0.099
0.090
0.093
0.097
0.088
0.102
0.096
P2O5
Total
0.240
99.508
0.241
99.003
0.237
99.083
0.235
99.192
0.243
99.051
0.295
98.448
0.207
98.613
0.204
99.590
0.255
99.358
0.241
98.537
0.243
98.627
0.231
99.406
0.244
98.896
0.241
98.390
0.237
99.254
0.243
98.962
0.237
98.596
0.220
99.288
0.228
99.181
0.190
99.65
0.177
99.85
0.171
99.45
0.164
99.81
0.175
99.80
0.167
99.70
0.169
99.56
0.174
98.806
0.075
99.603
0.073
99.68
0.113
99.28
0.139
98.80
0.135
99.52
0.176
99.47
0.173
99.85
0.176
99.64
0.218
99.07
0.187
99.36
0.206
98.96
0.170
99.33
0.163
99.30
0.168
99.52
0.188
99.59
0.153
99.86
0.299
99.39
0.222
99.42
0.259
100.28
0.159
100.14
0.249
99.87
0.239
99.79
0.196
100.12
0.205
100.84
0.179
100.17
0.206
99.54
0.290
99.76
0.250
99.62
0.187
99.73
0.203
99.86
0.094
100.09
0.117
99.89
0.102
100.33
0.096
100.33
0.203
99.19
0.167
99.63
0.205
99.50
0.214
99.64
0.203
99.34
0.128
99.37
Table 3. Gakkel Ridge SMZ and EVZ basalt sample major element concentration data
(wt.%), measured by electron microprobe. The distance column shows the distance along
the ridge from west to east, with the western end of the WVZ set as a distance of 0 km.
Magnesium number (Mg#) is defined as Mg#=MgO/40.3/(MgO+0.9*FeO/71.85), with
MgO and FeO denoting element concentrations. The total column displays the sum of all
of the major element concentrations. Although TiO2 alues were measured by electron
microprobe and reported in this table, the TiO2 values that were actually plotted in the
figures are the more precise values in Table 4, which were measured by LA-ICP-MS.
63
5.1.2) Gakkel Ridge trace element compositions
______________________________________________________________________________
Sample
D35-69
D35-38
D35-4
D35-2
D35-34
244-10
244-2
244-SG
D37-8
D36-53
D36-63
D36-66
D36-64
D36-70
D36-68
D36-59
D36-62
D36-2
D36-45
D38-26
D38-25
D38-27
D38-29R
D38-12
D38-4
D38-29
248-59
199-1-1
199-1-2
199-8
D41-18
D41-17
D42-16
D42-18
D42-12
252-1
253-1
254-1
255-1
255-3
255-5
315-1
D45-10
260-1
263-12
263-19
263-16
D47-7
309-5
265-2A
265-2
265-5
307-8
266-15
266-13
304-3
304-2
297-2
297-36
297-1
297-SG
271-1-11
271-1-9
270-2
270-19
295-2
294-2
Zone
Lon
Lat
Distance Rb
Sr
Y
Zr
Nb
Cs
Ba
La
Ce
Pr
Nd
Sm
Eu
Gd
SMZ 13
10.71 85.27
308
24
212
33.2
138
16 0.383
264
11.9
25.9
3.4
15.8
4.24
1.45
5.26
SMZ 13
10.71 85.27
308
15
178
34.3
129
14 0.212
163
9.8
22.5
3.1
14.7
4.08
1.44
5.21
SMZ 13
10.71 85.27
308
16
188
36.7
137
15 0.223
179
10.8
24.6
3.4
16.0
4.52
1.58
5.78
SMZ 13
10.71 85.27
308
15
186
35.9
136
14 0.225
177
10.6
24.3
3.3
16.0
4.54
1.54
5.66
SMZ 13
10.71 85.27
308
16
188
36.3
136
15 0.220
183
10.9
25.2
3.4
16.1
4.57
1.58
5.79
SMZ 13
11.04 85.05
312
32
304
27.1
120
26 0.369
371
16.8
33.5
3.9
16.8
3.72
1.34
4.33
SMZ 13
11.04 85.05
312
10
155
35.2
131
6 0.118
81.5
6.4
18.3
2.8
14.2
4.31
1.50
5.38
SMZ 13
11.04 85.05
312
10
155
36.0
132
6 0.143
81.9
6.6
18.3
2.8
14.1
4.02
1.49
5.42
SMZ 13
11.39 85.30
316
14
172
39.5
150
14 0.183
144
10.2
24.3
3.4
16.7
4.67
1.68
6.08
SMZ 13
12.48 85.27
329
24
215
32.6
136
17 0.384
268
11.9
26.3
3.4
15.7
4.19
1.43
5.09
SMZ 13
12.48 85.27
329
24
212
32.3
134
17 0.373
262
11.7
25.6
3.3
15.3
4.11
1.40
5.03
SMZ 13
12.48 85.27
329
24
215
33.7
139
16 0.371
265
12.0
25.6
3.4
15.5
4.14
1.41
5.11
SMZ 13
12.48 85.27
329
25
217
33.9
141
17 0.387
269
12.2
26.3
3.4
15.9
4.30
1.48
5.34
SMZ 13
12.48 85.27
329
24
213
32.5
136
16 0.365
265
12.0
26.0
3.4
15.6
4.18
1.40
4.98
SMZ 13
12.48 85.27
329
25
213
32.5
136
17 0.373
264
11.9
25.9
3.4
15.7
4.14
1.43
5.13
SMZ 13
12.48 85.27
329
24
211
32.5
135
16 0.370
255
11.4
25.0
3.3
15.0
4.03
1.37
4.90
SMZ 13
12.48 85.27
329
25
214
32.5
136
17 0.377
262
11.7
25.7
3.3
15.3
4.14
1.39
4.89
SMZ 13
12.48 85.27
329
23
209
32.1
135
16 0.373
257
11.4
25.1
3.3
15.4
4.01
1.39
4.95
SMZ 13
12.48 85.27
329
24
215
32.5
136
17 0.395
258
11.5
25.2
3.3
15.4
4.12
1.39
5.12
SMZ 13
12.68 85.31
332
8.8
160
31.5
113
8 0.111
92
6.5
16.2
2.4
11.9
3.67
1.28
4.70
SMZ 13
12.68 85.31
332
9.1
158
32.3
116
8 0.111
94
6.7
16.8
2.5
12.4
3.72
1.31
4.89
SMZ 13
12.68 85.31
332
8.0
152
29.6
105
7 0.102
84
6.1
15.1
2.2
11.3
3.43
1.25
4.51
SMZ 13
12.68 85.31
332
8.3
154
28.3
100
7 0.147
80
5.8
14.6
2.1
10.9
3.31
1.20
4.22
SMZ 13
12.68 85.31
332
7.9
152
29.8
104
7 0.098
83
6.0
15.0
2.2
11.2
3.42
1.19
4.47
SMZ 13
12.68 85.31
332
7.7
154
29.9
105
7 0.098
84
6.0
15.1
2.2
11.2
3.42
1.18
4.41
SMZ 13
12.68 85.31
332
8.0
154
29.6
104
7 0.101
85
6.0
15.2
2.2
11.3
3.43
1.23
4.54
SMZ 13
13.45 85.21
341
13
171
29.5
114
7 0.162
112
6.7
17.3
2.7
12.9
3.67
1.31
4.69
SMZ 19
16.21 85.56
371
0.7
139
25.2
63
1 0.044
9
2.1
6.8
1.2
6.9
2.40
1.03
3.63
SMZ 19
16.21 85.56
371
1.0
141
25.9
67
2 0.016
12
2.2
7.1
1.2
6.7
2.35
0.99
3.35
SMZ 19
16.21 85.56
371
9.0
172
22.9
79
5 0.143
84
4.6
11.7
1.7
8.3
2.51
0.98
3.32
SMZ 19
17.95 85.69
390
5.5
171
22.7
81
6 0.085
60
4.7
12.4
1.8
8.8
2.79
0.99
3.29
SMZ 19
17.95 85.69
390
5.7
182
23.5
88
6 0.070
64
4.9
12.7
1.9
9.4
2.86
1.05
3.77
SMZ 19
18.09 85.73
391
12
179
26.2
99
10 0.166
121
7.4
17.9
2.4
11.7
3.19
1.22
4.18
SMZ 19
18.09 85.73
391
12
180
25.9
97
10 0.168
119
7.2
17.8
2.4
11.5
3.22
1.16
4.00
SMZ 19
18.09 85.73
391
12
176
25.3
95
10 0.170
121
7.2
18.0
2.4
11.4
3.18
1.11
3.84
SMZ 19
18.22 85.62
392
19
217
25.8
114
14 0.280
198
9.8
22.2
2.9
13.2
3.43
1.20
4.10
SMZ 19
18.30 85.71
393
9.1
189
28.3
110
10 0.110
99
7.5
18.2
2.6
12.3
3.38
1.20
4.22
SMZ 19
19.50 85.74
405
8.7
201
25.5
101
10 0.108
102
7.3
17.6
2.4
11.6
3.20
1.16
4.07
SMZ 19
20.28 85.79
413
9.3
194
24.6
97
8 0.129
103
6.6
16.2
2.3
11.1
3.11
1.13
3.89
SMZ 19
20.28 85.79
413
9.3
195
24.5
97
8 0.131
103
6.5
15.8
2.2
10.7
3.03
1.10
3.70
SMZ 19
20.28 85.79
413
10
200
25.0
99
8 0.129
105
6.7
16.2
2.3
11.1
3.16
1.15
3.88
SMZ 19
20.45 85.81
415
8.7
171
31.0
109
8 0.154
81
6.6
16.9
2.4
12.8
3.71
1.34
4.60
SMZ 19
25.14 86.03
459
2.6
149
27.1
85
4 0.031
33
4.0
12.0
1.9
10.2
3.33
1.33
4.41
EVZ 31
29.14 85.97
494
7.9
137
51.9
178
11 0.083
98
9.3
23.9
3.7
19.1
5.96
1.97
7.99
EVZ 31
30.66 86.06
507
4.5
136
37.3
129
7 0.052
50
6.0
16.3
2.5
13.6
4.32
1.49
5.83
EVZ 31
30.66 86.06
507
6.5
157
40.2
145
10 0.069
78
7.8
19.7
3.0
15.6
4.85
1.66
6.49
EVZ 31
30.66 86.06
507
5.4
165
25.3
87
7 0.057
69
5.6
13.5
1.9
9.7
2.91
1.09
3.90
EVZ 31
31.27 86.05
512
6.9
158
36.1
136
10 0.082
79
8.1
21.7
3.1
15.9
4.73
1.66
5.93
EVZ 37
36.01 86.23
551
7.0
172
35.1
132
11 0.080
80
7.7
19.2
2.9
14.4
4.28
1.52
5.60
EVZ 37
37.75 86.33
565
4.3
144
38.2
122
7 0.048
54
5.9
15.7
2.5
13.4
4.30
1.54
5.84
EVZ 37
37.75 86.33
565
4.4
147
38.7
123
7 0.049
55
6.0
16.2
2.6
13.7
4.39
1.58
6.00
EVZ 37
37.75 86.33
565
3.8
176
26.8
103
6 0.059
44
5.3
14.1
2.2
11.2
3.23
1.21
4.24
EVZ 37
38.16 86.36
568
5.5
189
28.0
110
9 0.081
62
6.4
17.1
2.5
12.4
3.74
1.26
4.36
EVZ 37
38.76 86.31
573
8.9
164
40.1
154
13 0.114
97
9.1
23.5
3.5
17.2
5.15
1.66
6.31
EVZ 37
38.76 86.31
573
7.1
191
31.0
126
11 0.083
81
7.9
20.0
2.9
14.4
3.97
1.43
5.06
EVZ 43
40.99 86.48
591
4.0
169
31.0
118
6 0.053
45
5.7
15.8
2.4
12.2
3.49
1.33
4.77
EVZ 43
40.99 86.48
591
4.1
175
31.7
120
7 0.062
47
6.1
16.0
2.6
13.4
3.80
1.41
5.16
EVZ 55
47.11 86.68
639
1.3
157
24.9
64
3 0.041
15
2.6
7.24
1.1
6.2
2.09
0.87
3.15
EVZ 55
47.11 86.68
639
1.8
164
24.7
71
3 0.042
21
3.2
9.0
1.4
7.2
2.36
0.89
3.37
EVZ 55
47.11 86.68
639
1.4
162
24.6
64
3 0.031
16
2.7
7.8
1.3
6.7
2.23
0.94
3.22
EVZ 55
47.11 86.68
639
1.4
161
25.3
65
3 0.047
16
2.8
7.8
1.3
7.0
2.15
1.02
3.50
EVZ 55
47.72 86.84
644
5.1
202
24.5
102
8 0.069
56
6.0
15.7
2.3
11.3
3.08
1.20
3.96
EVZ 55
47.72 86.84
644
4.4
178
26.6
100
6 0.057
50
5.3
14.1
2.1
10.8
3.32
1.23
4.26
EVZ 55
50.12 86.75
663
5.1
240
26.6
117
9 0.065
61
7.5
18.3
2.7
13.3
3.55
1.31
4.49
EVZ 55
50.12 86.75
663
4.8
235
26.7
117
9 0.070
60
7.2
18.4
2.7
12.5
3.59
1.30
4.37
EVZ 55
52.30 86.80
681
4.7
193
27.5
111
8 0.066
56
6.6
16.6
2.4
12.2
3.40
1.24
4.38
EVZ 55
56.06 86.89
713
1.2
164
29.1
97
2 0.035
11
3.3
10.8
1.8
9.9
3.24
1.19
4.27
Table 4a. Gakkel Ridge SMZ and EVZ basalt sample trace element concentration data
(ppm), measured by LA-ICP-MS. The distance column shows the distance along the
ridge from west to east, with the western end of the WVZ set as a distance of 0 km.
64
______________________________________________________________________________
Sample
D35-69
D35-38
D35-4
D35-2
D35-34
244-10
244-2
244-SG
D37-8
D36-53
D36-63
D36-66
D36-64
D36-70
D36-68
D36-59
D36-62
D36-2
D36-45
D38-26
D38-25
D38-27
D38-29R
D38-12
D38-4
D38-29
248-59
199-1-1
199-1-2
199-8
D41-18
D41-17
D42-16
D42-18
D42-12
252-1
253-1
254-1
255-1
255-3
255-5
315-1
D45-10
260-1
263-12
263-19
263-16
D47-7
309-5
265-2A
265-2
265-5
307-8
266-15
266-13
304-3
304-2
297-2
297-36
297-1
297-SG
271-1-11
271-1-9
270-2
270-19
295-2
294-2
Zone
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 13
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
SMZ 19
EVZ 31
EVZ 31
EVZ 31
EVZ 31
EVZ 31
EVZ 37
EVZ 37
EVZ 37
EVZ 37
EVZ 37
EVZ 37
EVZ 37
EVZ 43
EVZ 43
EVZ 55
EVZ 55
EVZ 55
EVZ 55
EVZ 55
EVZ 55
EVZ 55
EVZ 55
EVZ 55
EVZ 55
Lon
Lat
Distance Tb
Dy
Ho
Er
Tm
Yb
Lu
Hf
Ta
Pb
Th
U
10.71 85.27
308
0.89
5.89
1.23
3.57
0.509
3.5 0.520
3.28
0.943
1.06
1.58 0.373
10.71 85.27
308
0.88
5.74
1.21
3.52
0.501
3.4 0.482
3.01
0.742
0.90
1.09 0.266
10.71 85.27
308
0.98
6.50
1.37
4.01
0.571
3.9 0.571
3.50
0.859
1.03
1.28 0.309
10.71 85.27
308
0.98
6.51
1.34
3.88
0.564
3.8 0.563
3.38
0.828
0.99
1.26 0.310
10.71 85.27
308
0.98
6.43
1.36
3.96
0.579
3.9 0.573
3.44
0.855
1.05
1.26 0.316
11.04 85.05
312
0.73
4.65 0.971
2.69
0.408
2.7 0.385
2.76
1.290
1.40
2.14 0.466
11.04 85.05
312
0.91
6.10
1.26
3.58
0.540
3.6 0.510
3.08
0.359
0.82
0.52 0.124
11.04 85.05
312
0.97
6.48
1.31
3.70
0.535
3.8 0.540
3.11
0.376
0.87
0.56 0.170
11.39 85.30
316
1.0
6.81
1.45
4.24
0.591
4.0 0.596
3.59
0.791
1.04
1.09 0.275
12.48 85.27
329
0.87
5.83
1.20
3.57
0.508
3.5 0.509
3.26
0.948
1.08
1.57 0.353
12.48 85.27
329
0.86
5.70
1.17
3.52
0.497
3.4 0.501
3.17
0.904
1.04
1.52 0.330
12.48 85.27
329
0.89
5.87
1.21
3.56
0.522
3.4 0.526
3.22
0.923
1.10
1.58 0.338
12.48 85.27
329
0.90
6.00
1.23
3.67
0.528
3.6 0.530
3.34
0.977
1.12
1.59 0.342
12.48 85.27
329
0.86
5.66
1.18
3.55
0.491
3.4 0.503
3.10
0.914
1.02
1.54 0.340
12.48 85.27
329
0.88
5.91
1.19
3.57
0.506
3.5 0.509
3.24
0.936
1.08
1.55 0.347
12.48 85.27
329
0.84
5.64
1.15
3.44
0.482
3.3 0.483
3.05
0.898
1.02
1.48 0.328
12.48 85.27
329
0.85
5.66
1.15
3.49
0.490
3.4 0.487
3.12
0.917
1.08
1.53 0.331
12.48 85.27
329
0.85
5.49
1.16
3.43
0.489
3.3 0.487
3.09
0.911
1.02
1.50 0.346
12.48 85.27
329
0.85
5.56
1.17
3.44
0.485
3.3 0.479
3.05
0.900
0.99
1.48 0.347
12.68 85.31
332
0.82
5.46
1.14
3.47
0.479
3.3 0.490
2.78
0.489
0.76
0.65 0.152
12.68 85.31
332
0.84
5.61
1.18
3.50
0.496
3.4 0.502
2.84
0.509
0.77
0.67 0.155
12.68 85.31
332
0.78
5.11
1.06
3.16
0.447
3.1 0.452
2.54
0.455
0.71
0.58 0.131
12.68 85.31
332
0.76
5.03
1.01
2.96
0.429
3.0 0.435
2.45
0.441
0.66
0.60 0.157
12.68 85.31
332
0.77
5.18
1.07
3.22
0.452
3.1 0.460
2.58
0.444
0.69
0.58 0.133
12.68 85.31
332
0.76
5.02
1.08
3.13
0.449
3.1 0.444
2.53
0.448
0.68
0.58 0.136
12.68 85.31
332
0.76
5.17
1.09
3.17
0.455
3.1 0.471
2.65
0.463
0.76
0.59 0.136
13.45 85.21
341
0.81
5.32
1.12
3.14
0.477
3.2 0.471
2.86
0.444
0.85
0.60 0.148
16.21 85.56
371
0.64
4.64 0.976
2.91
0.436
2.9 0.433
1.68
0.083
0.35
0.10 0.038
16.21 85.56
371
0.63
4.33 0.941
2.83
0.409
2.8 0.419
1.70
0.098
0.34
0.11 0.033
16.21 85.56
371
0.59
4.03 0.824
2.51
0.344
2.4 0.360
1.86
0.326
0.58
0.43 0.113
17.95 85.69
390
0.59
3.91 0.786
2.45
0.337
2.4 0.347
2.03
0.327
0.58
0.40 0.129
17.95 85.69
390
0.63
4.25 0.886
2.56
0.359
2.5 0.371
2.18
0.345
0.62
0.43 0.103
18.09 85.73
391
0.70
4.74 0.959
2.93
0.419
2.8 0.431
2.37
0.580
0.79
0.78 0.219
18.09 85.73
391
0.70
4.65 0.941
2.82
0.415
2.9 0.415
2.37
0.562
0.78
0.79 0.216
18.09 85.73
391
0.66
4.45 0.943
2.92
0.410
2.9 0.404
2.07
0.589
0.80
0.77 0.219
18.22 85.62
392
0.68
4.42 0.922
2.76
0.391
2.7 0.390
2.59
0.824
1.02
1.20 0.289
18.30 85.71
393
0.70
4.68 0.978
2.96
0.418
2.8 0.413
2.48
0.567
0.76
0.72 0.226
19.50 85.74
405
0.69
4.52 0.934
2.79
0.393
2.7 0.386
2.34
0.573
0.71
0.74 0.194
20.28 85.79
413
0.68
4.45 0.906
2.73
0.379
2.6 0.378
2.28
0.495
0.74
0.66 0.165
20.28 85.79
413
0.64
4.41 0.886
2.65
0.380
2.6 0.371
2.30
0.494
0.71
0.67 0.162
20.28 85.79
413
0.66
4.49 0.911
2.71
0.378
2.6 0.373
2.27
0.491
0.73
0.67 0.169
20.45 85.81
415
0.84
5.56
1.12
3.42
0.492
3.1 0.481
2.83
0.497
0.90
0.63 0.356
25.14 86.03
459
0.76
5.06
1.02
3.12
0.440
3.0 0.420
2.37
0.275
0.58
0.33 0.105
29.14 85.97
494
1.4
9.08
1.91
5.66
0.811
5.5 0.814
4.48
0.699
1.02
0.83 0.278
30.66 86.06
507
1.0
6.62
1.38
4.09
0.576
3.9 0.581
3.24
0.425
0.68
0.52 0.157
30.66 86.06
507
1.1
7.17
1.50
4.43
0.623
4.2 0.635
3.76
0.631
0.81
0.72 0.199
30.66 86.06
507
0.67
4.39 0.913
2.70
0.377
2.6 0.381
2.10
0.409
0.60
0.49 0.141
31.27 86.05
512
1.0
6.74
1.38
4.11
0.589
4.1 0.589
3.68
0.620
0.95
0.77 0.234
36.01 86.23
551
0.96
6.29
1.31
3.81
0.530
3.7 0.535
3.33
0.646
0.80
0.75 0.219
37.75 86.33
565
1.0
6.78
1.41
4.17
0.585
4.0 0.590
3.21
0.429
0.68
0.50 0.139
37.75 86.33
565
1.0
6.83
1.43
4.25
0.606
4.1 0.604
3.24
0.440
0.71
0.50 0.147
37.75 86.33
565
0.72
4.63
1.01
2.94
0.381
2.8 0.408
2.42
0.350
0.63
0.41 0.119
38.16 86.36
568
0.76
4.94
1.01
3.01
0.413
2.7 0.417
2.58
0.491
0.70
0.52 0.146
38.76 86.31
573
1.1
6.82
1.46
4.25
0.575
4.0 0.587
3.77
0.719
0.91
0.85 0.234
38.76 86.31
573
0.85
5.45
1.19
3.35
0.461
3.2 0.473
3.04
0.616
0.85
0.72 0.211
40.99 86.48
591
0.81
5.40
1.15
3.26
0.448
3.1 0.455
2.80
0.387
0.67
0.44 0.126
40.99 86.48
591
0.87
5.74
1.25
3.51
0.489
3.3 0.490
3.08
0.400
0.69
0.44 0.137
47.11 86.68
639
0.56
4.05 0.873
2.70
0.378
2.7 0.428
1.51
0.172
0.34
0.17 0.057
47.11 86.68
639
0.60
3.94 0.902
2.75
0.388
2.7 0.424
1.66
0.215
0.36
0.22 0.088
47.11 86.68
639
0.63
4.15 0.918
2.76
0.416
2.8 0.435
1.61
0.174
0.36
0.18 0.057
47.11 86.68
639
0.63
4.38 0.980
2.90
0.419
3.1 0.454
1.71
0.180
0.35
0.18 0.064
47.72 86.84
644
0.66
4.23 0.905
2.63
0.350
2.5 0.370
2.39
0.495
0.68
0.51 0.180
47.72 86.84
644
0.71
4.79
1.00
2.83
0.404
2.9 0.408
2.44
0.389
0.62
0.42 0.147
50.12 86.75
663
0.71
4.78
1.00
2.98
0.390
2.9 0.410
2.70
0.588
0.79
0.59 0.202
50.12 86.75
663
0.71
4.66 0.981
2.89
0.400
2.6 0.396
2.65
0.550
0.76
0.57 0.189
52.30 86.80
681
0.74
4.78
1.03
2.98
0.408
3.0 0.420
2.63
0.501
0.71
0.54 0.167
56.06 86.89
713
0.73
4.88
1.03
3.11
0.447
3.1 0.448
2.37
0.167
0.55
0.16 0.057
TiO2 Sc
1.52
33.8
1.65
34.4
1.67
35.2
1.67
34.7
1.69
35.4
1.63
31.4
1.62
36.7
1.67
36.3
1.80
34.5
1.58
33.5
1.58
33.8
1.56
34.6
1.57
34.0
1.58
34.1
1.59
33.9
1.58
33.8
1.57
33.4
1.53
33.6
1.55
33.6
1.43
34.7
1.49
35.9
1.40
34.9
1.38
34.1
1.40
35.0
1.40
35.1
1.39
34.7
1.30
34.1
1.18
40.5
1.21
39.5
1.18
33.7
1.12
36.5
1.14
37.0
1.34
35.6
1.35
35.9
1.33
34.9
1.39
31.1
1.43
38.3
1.24
34.3
1.21
37.4
1.20
37.0
1.22
37.6
1.50
36.9
1.44
32.1
2.30
41.0
1.75
36.3
2.04
38.6
1.35
35.2
1.76
37.6
1.76
36.0
1.74
39.8
1.79
40.0
1.43
34.7
1.54
33.1
1.98
34.2
1.68
34.2
1.52
34.8
1.56
36.3
1.05
40.7
1.10
38.5
1.03
40.2
1.03
40.2
1.46
32.6
1.42
34.8
1.59
35.3
1.52
34.0
1.44
34.9
1.37
35.6
Table 4b. Gakkel Ridge SMZ and EVZ basalt sample trace element concentration data
(ppm), measured by LA-ICP-MS (continued from Table 4a). The major element TiO2
was also measured by LA-ICP-MS and its values are reported in this table.
65
5.2)
Ternary plots – Arctic MORBs
Figure 12a. NKL ternary plot showing relative concentrations of Na2O (wt.% x 5), K2O
(wt.% x 30), & La (ppm) concentrations in Arctic Basin Gakkel Ridge and Lena Trough
MORB glass samples. Uncertainty polygons are shown as black dots surrounding the
sample points analyzed in this study. Samples that were not analyzed in this study do not
have uncertainty polygons. Gakkel Ridge data are a combination of data from this study
and from previously published data (Michael et al., 2003; Goldstein et al., 2008; Shaw et
al., 2010). Also plotted are CLT basalt data (Nauret et al., 2011), WVZ & NLT (Snow
J.E., in prep), PM (Sun & McDonough, 1989), and DMM (Salters & Stracke, 2004).
Average N-MORB and E-MORB are calculated from basalt glass data from PetDB
(Lehnert et al., 2000). The global MORB array, comprising data from the MAR, EPR,
SWIR, & SEIR, (PetDB; Lehnert et al., 2000) is encircled in gray. Sample 297-2 from
EVZ55 and its uncertainty points (from the YSL ternary plot) are shown in the legend to
represent a typical analytical uncertainty for Gakkel samples.
66
Figure 12b. TKL ternary plot showing relative concentrations of TiO2 (wt.% x 5), K2O
(wt.% x 30), & La (ppm) in Gakkel Ridge and Lena Trough basalt glasses. Uncertainty
polygons are shown as black dots surrounding the sample points analyzed in this study.
The global MORB array is encircled in gray. Sample 297-2 from EVZ55 and its
uncertainty points (from the YSL ternary plot) are shown in the legend to represent a
typical analytical uncertainty for Gakkel samples. Data are from the same references as
those in Figure 12a.
67
Figure 12c. YSL ternary plot showing relative Yb (ppm), Sm (ppm x 2), & La (ppm)
concentrations in Gakkel Ridge and Lena Trough basalt glasses. Uncertainty polygons
are shown as black dots surrounding the sample points analyzed in this study. The global
MORB array is encircled in gray. Sample 297-2 from EVZ55 and its uncertainty points
(from the YSL ternary plot) are shown in the legend to represent a typical analytical
uncertainty for Gakkel samples. Data are from the same references as those in Figure 12a.
Figure 12 shows trends in the Arctic Basin basalt composition data when plotted
onto ternary diagrams. The data appear split into two separately trending data arrays. One
array contains WVZ, NLT, and CLT data, while the other array contains SMZ, EVZ, and
most of the global MORB data. WVZ*’s plotting position makes it seem like it may be
68
the depleted endmember of the data array trending between WVZ-CLT, while CLT
appears to be its enriched endmember. WVZ* is distinguished from WVZ samples by
their characteristic depletion in certain trace elements (Goldstein et al., 2008). SMZ and
EVZ data, instead of trending towards the unusual CLT composition, appear to plot more
in line with the general depleted-to-enriched trend of most global MORBs.
6)
DISCUSSION
In this study, trace element concentrations were measured in Gakkel Ridge basalt
glass samples to determine the number of source components in the mantle along Gakkel
Ridge and the nature of those components. To accomplish this, major element and trace
element concentrations were measured in basalt glass samples recovered from the
SMZ13, SMZ19, EVZ31, EVZ37, EVZ43, and EVZ55 volcanic centers. The newly
acquired major and trace element concentration data were combined with previously
published Gakkel Ridge major and trace element concentration data (Michael et al., 2003;
Goldstein et al., 2008; Shaw et al., 2010) and plotted on ternary plots (Fig. 12) to
compare relative differences between sets of trace elements. Interpretations regarding the
nature of the mantle source components along Gakkel Ridge were made based on these
diagrams.
6.1)
Enriched component reservoirs
On the ternary plots produced for this study, basalt sample trace element
composition data points that plot within each other’s fields of uncertainty are likely to
have formed from the identical magma compositions. However, samples that do not plot
69
within uncertainty of each other must have cooled from magmas with different
compositions. If these basalts plot as an array between two endpoints, then it’s likely an
indication that those basalts are the products of mixing between two endmember source
component compositions. The chemical compositions of those endmember components
can be interpreted to resemble the compositions represented at the endpoints of the twocomponent mixing lines. Such interpretations can be made for basalt samples from
ultraslow spreading ridges such as Gakkel Ridge or Lena Trough, because ultraslow
spreading rates can minimize the extent of the fractional crystallization process.
The most significant mineral phase to crystallize at such low degrees of
crystallization is olivine, which is a mineral into which many trace elements are
extremely incompatible, so the trace element composition of the magma changes very
little as it crystallizes (Fig. 4a-4b). Crystallization of additional phases, such as
clinopyroxene (Fig. 4c-4d), generate more significant changes in magma trace element
compositions than during olivine crystallization (Fig. 4). But at ultraslow spreading
ridges, the extent to which these other phases can crystallize would be insignificant
(Walker, 1979; Stolper, 1980), so their effects on magma compositions are negligible.
The minimal crustal thickness associated with ultraslow spreading, such as at Gakkel
Ridge, where the crust is especially thin (Jokat et al., 2003), further implies that
clinopyroxene crystallization would be irrelevant at Gakkel Ridge. Also, it’s notable that
even at the calculated 50% crystallization of clinopyroxene (Fig. 4c-4d), the effect on the
diagram is still close to analytical uncertainty. As a consequence of all these factors,
MORB samples from ultraslow spreading ridges, including Gakkel Ridge and Lena
70
Trough, and the magma compositions corresponding to those basalt compositions end up
plotting nearly identically on trace element ternary plots, as demonstrated by Figure 4.
6.1.1) Enriched reservoirs: size & distribution
The uncertainty of the data points plotted on the trace element concentration
ternary plots were calculated by adding and subtracting the calculated standard deviations
for each trace element concentration measurement plotted on each ternary plot, forming
six-sided uncertainty polygons surrounding each plotted data point (Fig. 13). The
uncertainties of the elements of the YSL ternary plot were all calculated from the
standard deviations of the analyses of those elements from the analyses of the basalt
standard BIR-1.
Uncertainty polygons are shown for samples of the SMZ13 (Fig. 13a), SMZ19
(Fig. 13b), EVZ31, EVZ37, EVZ43 (Fig. 13c), and EVZ55 (Fig. 13d) volcanic centers on
YSL ternary plots. The size of the uncertainty polygons correlate with the level of
precision associated with the trace element concentration measurements for each basalt
sample data point. Smaller uncertainty polygons (e.g., sample 199-1-2, Fig. 10) represent
less uncertainty and greater precision, while larger uncertainty polygons (e.g., sample
252-1, Fig. 10) represent more uncertainty and less precision.
Basalt samples, recovered from a distance spanning 97 kilometers along Gakkel
Ridge across three volcanic centers EVZ31, EVZ37, and EVZ43, all plot relatively close
to each other, with many points plotting within each other’s uncertainty polygons on the
YSL ternary plot (Fig. 13c). In contrast, basalt samples, recovered from the SMZ13 (Fig.
13a), SMZ19 (Fig. 13b), and EVZ55 (Fig. 13d) volcanic centers, plot relatively far apart
71
from each other on the YSL ternary plot, with many samples plotting outside of each
other’s uncertainty polygons. The large amount of trace element variations within the
SMZ13, SMZ19, and EVZ55 volcanic centers, with many samples plotting outside of
each other’s uncertainty fields (Fig. 13a; Fig. 13b; Fig. 13d), indicate a relatively large
degree of variation in the size and distribution of the enriched reservoirs in the mantle
beneath each of these volcanic centers. These enriched heterogeneities must be at least
small enough and distributed widely enough to produce significant trace element
variations between individual sample dredging stations.
72
Figure 13a. Uncertainty polygons surrounding the data points of SMZ13 basalt samples
plotted on an NKL ternary plot comparing Na2O (wt.% x 5), K2O (wt.% x 30), & La
(ppm). The filled gray area encompasses the Gakkel Ridge data set from this study. The
global MORB array, comprising data from the MAR, EPR, SWIR, & SEIR, (PetDB;
Lehnert et al., 2000) is encircled in gray. Average N-MORB and E-MORB are calculated
from basalt glass data from PetDB (Lehnert et al., 2000). Data are from this study and
earlier studies (Michael et al., 2003; Goldstein et al., 2008). SMZ13 samples are
recovered from a span of 33 kilometers along the ridge (Fig. 7).
73
Figure 13b. Uncertainty polygons surrounding the data points of NKL ternary plot
comparing Na2O (wt.% x 5), K2O (wt.% x 30), & La (ppm). The filled gray area
encompasses the Gakkel Ridge data set from this study. The global MORB array is
encircled in gray. Data are from the same references as those in Figure 13a. SMZ19
samples are recovered from a span of 87 kilometers along the ridge (Fig. 7).
74
Figure 13c. Uncertainty polygons surrounding the data points of EVZ31, EVZ37, and
EVZ43 basalt samples plotted a NKL ternary plot comparing Na2O (wt.% x 5), K2O
(wt.% x 30), & La (ppm). The filled gray area encompasses the Gakkel Ridge data set
from this study. The global MORB array is encircled in gray. Data are from the same
references as those in Figure 13a. EVZ31, EV37, and EVZ43 basalts were all recovered
from a distance spanning 97 kilometers.
The EVZ31, EVZ37, and EVZ43 basalt samples, despite spanning a relatively
long distance of 97 kilometers along Gakkel Ridge, all plot relatively closer together,
with several samples plotting within range of each other’s uncertainties (Fig. 13c). Such a
signature shows that the magmatic mixing behavior along the entire length across the
75
EVZ31, EVZ37, and EVZ43 volcanic centers is relatively consistent. This may be
interpreted to indicate that there is relatively limited variation in the size and distribution
of the enriched reservoirs along this length of Gakkel Ridge.
Figure 13d. Uncertainty polygons surrounding the data points of EVZ55 basalt samples
plotted on a YSL ternary plot comparing La (ppm), Sm (ppm x 2), and Yb (ppm).
Samples from dredging station 297 (AMORE 2001) are circled in red The filled gray area
encompasses the Gakkel Ridge data set from this study. The global MORB array is
encircled in gray. Data are from the same references as those in Figure 13a. EVZ55
samples are recovered from a span of 74 kilometers along the ridge (Fig. 7).
76
The plotted trace element signature in basalts from EVZ55 does not simply follow
one mixing line from one depleted point to one enriched point, but instead, features
samples plotting along at least two different trajectories (Fig. 13d). Such a pattern can
only be produced by mixing between at least three distinct source component
compositions. This means that in addition to the enriched component resembling MAR
E-MORB, at least one other enriched component must exist along Gakkel Ridge.
Two of the components present along the SMZ and EVZ magmatic zones of
Gakkel Ridge are the DMM component and an enriched component resembling MAR EMORB. The third component is distinct in its trace element chemistry from both the
DMM and MAR E-MORB-like components and cannot be produced by magmatic
mixing between melts derived from those two components, because its composition does
not plot anywhere in between those two components. This is indicated most clearly by
the chemical signature of the basalts of the EVZ55 volcanic center, which display on the
ternary plots a trace element pattern that can only be produced by at least three different
source component compositions, because the plotted samples do not follow a single
trajectory between two endpoints, but instead appear to plot along at least two different
trajectories (Fig. 13d). The influence of the third component is exemplified by samples
297-1, 297-2, 297-36, and 297-SG (circled in red in Fig. 13d), all of which were dredged
from station 297, which is located at the western end of the volcanic center EVZ55 (Fig.
8).
77
6.1.2) Enriched component: origin
As a result of the limited effect of melt differentiation processes on melt
compositions (Fig. 4), basalt compositions and their respective mantle melting source
component reservoir compositions should plot so that they overlap each other and should
be nearly indistinguishable from each other on trace element composition ternary plots
(Fig. 3; Fig. 5; Fig. 12). This means that basalt compositions plotted on trace element
concentration ternary plots are representative of both the basalt composition and the
magma from which that basalt was derived. Knowing this, two-component mixing lines
can be drawn on the ternary plots and mantle endmember source component
compositions can be determined. The origins of those enriched components can then be
interpreted.
Two-component mixing lines were calculated and plotted on the TKL ternary plot
(Fig. 14). CLT, NLT, and WVZ data plot along a mixing line between DMM and CLT.
The SMZ and EVZ data array plots along a mixing line between DMM and average
MAR E-MORB (Fig. 14c). The SMZ and EVZ trace element concentration data array
plots closer to the mixing line between DMM and EM1 (Fig. 14b) than the mixing line
between DMM and EM2 (Fig. 14a).
78
Figure 14a. TKL ternary plot, showing La (ppm), K2O (wt.% x 30), & TiO2 (wt.% x 5),
displaying a mixing line, colored red, between DMM (Salters & Stracke, 2004) and the
extreme-K2O CLT sample (Nauret et al., 2011) and a separate mixing line, colored green,
between the DMM and the averaged EM2 composition, calculated by combining EM2
OIB whole rock and glass concentration data from the Samoa and Societies islands (EM2
data are compiled from GEOROC (Lehnert et al., 2000)). Average N-MORB and EMORB are calculated from basalt glass data from PetDB (Lehnert et al., 2000). The
global MORB array, comprising data from the MAR, EPR, SWIR, & SEIR, (PetDB;
Lehnert et al., 2000) is encircled in gray.
79
Figure 14b. TKL ternary plot showing La (ppm), K2O (wt.% x 30), & TiO2 (wt.% x 5),
displaying a mixing line, colored red, between DMM (Salters & Stracke, 2004) and the
extreme-K2O CLT sample (Nauret et al., 2011) and a separate mixing line, colored blue,
between the DMM and the averaged EM1 composition, calculated by combining EM1
OIB whole rock and glass concentration data from the Austral-Cook, St. Helena, Gough,
Kerguelen, Pitcairn-Gambier, Tristan da Cunha, and Walvis Ridge islands (EM1 data
compiled from GEOROC (Lehnert et al., 2000)). Average N-MORB and E-MORB are
calculated from basalt glass data from PetDB (Lehnert et al., 2000). The global MORB
array, comprising data from the MAR, EPR, SWIR, & SEIR, (PetDB; Lehnert et al.,
2000) is encircled in gray.
80
Figure 14c. TKL ternary plot showing La (ppm), K2O (wt.% x 30), & TiO2 (wt.% x 5),
displaying a mixing line, colored red, between DMM (Salters & Stracke, 2004) and the
extreme-K2O CLT sample (Nauret et al., 2011) and a separate mixing line, colored blue,
between the DMM and the averaged MAR E-MORB composition. MAR basalt glass data
compiled from PetDB (Lehnert et al., 2000). MAR E-MORB is defined as all MAR
MORB samples with a La/Sm ratio greater than 1.8. The global MORB array, comprising
data from the MAR, EPR, SWIR, & SEIR, (PetDB; Lehnert et al., 2000) is encircled in
gray.
The averaged compositions of basalts derived from the enriched OIB components,
EM1, EM2, and HIMU, were compared to the enrichment in the SMZ and EVZ, because
81
the origins of the enriched OIB components have been thoroughly interpreted and
described (Weaver 1991a; Weaver 1991b). If the enriched component in the SMZ and
EVZ resemble any of the OIB enriched components, it may mean that they share the
same origin. The EM2 chemical signature is said to be indicative of a continental crust
component in the mantle (Weaver, 1991a; Weaver 1991b; Hofmann, 1997). The EM1
and HIMU chemical signatures are said to be indicative of recycled oceanic crust
(Weaver, 1991a; Weaver 1991b). The data array on which SMZ and EVZ basalt trace
element data plot, which trends between DMM and average MAR E-MORB on the trace
element concentration ternary plots (Fig. 14), can be compared to the two-component
mixing lines between DMM and average EM1 or between DMM and average EM2 to see
whether EM1 or EM2 – thus, whether recycled oceanic crust or recycled continental
crust, respectively – is the origin for the enriched component at the SMZ and EVZ. Most
HIMU basalt trace element concentration data clearly plot off the trend between DMM
and average MAR E-MORB (Fig. 5; Fig. 14), so a mixing line between DMM and
average HIMU was not calculated and plotted. The mixing line between DMM and
average MAR E-MORB (Fig. 14c) appears to be more similar in its trending direction to
the mixing line between DMM and average EM1 (Fig. 14b) than the mixing line between
DMM and average EM2 (Fig. 14a). This appears to suggest that recycled oceanic crust is
a more likely origin for the SMZ and EVZ enriched component than recycled continental
crust. This result is consistent with previous studies, which concluded that no signature of
any continental mantle has been found in MORB samples of eastern Gakkel Ridge (Liu et
al., 2008).
82
A Spitsbergen SCLM component has been suggested to be a source of enrichment
influencing western Gakkel Ridge (Goldstein et al., 2008), so Spitsbergen basalts were
considered to be worth examining as a possible source of enrichment in the SMZ and
EVZ. The influence of the Spitsbergen SCLM component on Gakkel Ridge is visible on
basalt isotope concentration plots (Fig. 10a), which show Spitsbergen basalts plotting
along the same trend as Lena Trough and western Gakkel Ridge samples. However,
although the Spitsbergen isotope data have been published (Ionov, 2002), the trace
element data required to compare Spitsbergen samples with Gakkel Ridge samples are
not available, so the Spitsbergen samples could not be included on the trace element
figures in this study. The Spitsbergen SCLM being present in eastern Gakkel Ridge is
unlikely, however, because according to Liu et al. (2008) no signature of any continental
mantle in the mantle source of eastern Gakkel Ridge has been found.
Another enriched component considered as a possible source of enrichment for
the SMZ and EVZ is the unique K-enriched CLT component, which has been described
to possibly be SCLM in origin (Nauret et al., 2011). Based on the two-component mixing
lines calculated for the trace element ternary plots however, the CLT component does not
influence the trace element enrichment of the SMZ and EVZ, despite the CLT component
being a clear influence on the trace element enrichment of the WVZ and NLT (Fig. 14).
This distinction between the DMM-CLT mixing trend, on which the WVZ basalt trace
element data plot, and the DMM-MAR E-MORB trend, on which the SMZ and EVZ
basalt trace element data plot, is shown on Figure 14. This means that the k-enriched
CLT component, which likely has a SCLM origin (Nauret et al., 2011), is not the same
enriched component present in the SMZ and EVZ.
83
6.2)
Isotopic boundary
If the isotopic boundary defined by Goldstein et al. (2008) is also reflected in
trace elements, then SMZ13 samples might be expected to plot on the same mixing trend
as WVZ, which trends towards CLT on trace element diagrams. However, the isotopic
boundary apparently does not correlate with the observable trends seen on trace element
ternary plots (Fig. 14), which show SMZ13 samples plotting along a more similar trend
to SMZ19 and EVZ samples in the eastern part of Gakkel Ridge, instead of along with
the WVZ and Lena Trough samples in the western part of Gakkel Ridge. There is an
apparent inconsistency between the location of the isotopic boundary suggested by
Goldstein et al. (2008), located in the middle of the SMZ, and the separation of the two
mixing lines observable on the trace element ternary plots (Fig. 14), which occurs in
between the WVZ and SMZ volcanic zones. The reason why the isotopic boundary
identified by Goldstein et al. (2008) does not manifest itself in the trace element signature
along Gakkel Ridge is not clear.
7)
CONCLUSIONS
The following conclusions about the mantle melting source components along
Gakkel Ridge were made based on analysis of the major and trace element and isotope
data:
1) The enriched mantle source component in the SMZ and EVZ is similar in
chemical composition to MAR E-MORB. This is suggested by the similarities in trends
between the SMZ and EVZ data array on trace element composition ternary plots to the
calculated two-component mixing line between DMM and average MAR E-MORB. The
84
enriched component in the SMZ and EVZ is chemically distinct from the CLT
component, which is present as a mixing component in the NLT and WVZ, but not in the
SMZ and EVZ.
2) The MAR E-MORB-like enriched mixing component in the SMZ and EVZ is
not likely to be subducted continental crust in origin, because the SMZ-EVZ data array
does not correlate with the two-component mixing line between the DMM and EM2 on
the trace element ternary plots (Fig. 14a), which has been suggested to have a subducted
continental crust origin.
3) The two-component mixing line between DMM and average EM1 is close in
trend to the mixing line between DMM and average EM1. The source of enrichment for
EM1 has been suggested to be recycled oceanic crust in origin, so the origin of the main
enriched SMZ and EVZ component may also be recycled oceanic crust.
4) In addition to the enriched component resembling MAR E-MORB, at least one
other enriched composition exists along Gakkel Ridge. This additional enriched
component is most clearly identifiable from the EVZ55 trace element signature, which
shows a pattern that can only be produced from melts derived at least three different
endmember source compositions.
5) The level of variation in trace element enrichment between volcanic centers
along Gakkel Ridge is minimal along a lengthy section of Gakkel Ridge. Samples
collected from a 97 kilometer length spanning the EVZ31, EVZ37, and EVZ43 volcanic
centers show relatively little trace element variation between samples, with all samples
mostly plotting relatively close to each other, with many samples plotting within
uncertainty of each other on the trace element concentration ternary plots. The limited
85
trace element variation across these volcanic centers suggests that the size and
distribution of the enriched reservoirs beneath Gakkel Ridge across this entire length is
relatively consistent. This results in consistent mixing behavior between enriched and
depleted magmas, so that the basalts produced have relatively similar compositions along
that entire length along Gakkel Ridge.
6) The trace element variation within the each of the three volcanic centers
SMZ13, SMZ19, and EVZ55 is significantly greater than the minimal variation
observable across the three volcanic centers EVZ31, EVZ37, and EVZ43. This is despite
each of the volcanic centers SMZ13, SMZ19, and EVZ55 spanning a distance that is less
than the combined distance spanned by all three of the EVZ31, EVZ37, and EVZ43
volcanic centers. The large amount of trace element variation that exists between samples
from within the SMZ13, SMZ19, and EVZ55 volcanic centers suggests that the enriched
reservoirs in the mantle beneath these volcanic centers are small enough and distributed
with enough spacing in between them to produce significant trace element variation in
mixed magmas between distances as short as the distances in between dredging stations.
86
8)
APPENDIX
8.1)
Appendix 1: XY plots
X-Y plots showing Gakkel Ridge trace element ratios plotted against magnesium
oxide (MgO), magnesium number (Mg#), and longitude. Gakkel Ridge data include new
data from this study, and previously published data (Michael et al., 2003; Goldstein et al.,
2008). Knipovich data compiled from PetDB (Lehnert et al., 2000). Macquarie Island
data (Kamenetsky et al., 2000) is included as a representative E-MORB composition.
MgO, K2O, and TiO2 concentrations are in wt.%; Ba, Dy, Yb, La, Sm, and Nb
concentrations are in ppm.
The X-Y plots are:
Fig. A1a: Ba/TiO2 vs. MgO
Fig. A1b: Dy/Yb vs. MgO
Fig. A1c: K2O/La vs. MgO
Fig. A1d: K2O/TiO2 vs. MgO
Fig. A1e: La/Sm vs. MgO
Fig. A1f: Nb/La vs. MgO
Fig. A1g: Ba/TiO2 vs. Mg#
Fig. A1h: Dy/Yb vs. Mg#
Fig. A1i: K2O/La vs. Mg#
Fig. A1j: K2O/TiO2 vs. Mg#
Fig. A1k: La/Sm vs. Mg#
Fig. A1l: Nb/La vs. Mg#
Fig. A1m: Ba/TiO2 vs. Longitude
Fig. A1n: Dy/Yb vs. Longitude
Fig. A1o: K2O/La vs. Longitude
Fig. A1p: K2O/TiO2 vs. Longitude
Fig. A1q: La/Sm vs. Longitude
Fig. A1r: Nb/La vs. Longitude
87
Figure A1a. Comparison of Ba/TiO2 (ppm) to MgO (wt.%).
Figure A1b. Comparison of Dy/Yb (ppm) to MgO (wt.%).
88
Figure A1c. Comparison of K2O/La to MgO (wt.%). K2O is in wt.%. La is in ppm.
Figure A1d. Comparison of K2O/TiO2 (wt.%) to MgO (wt.%).
89
Figure A1e. Comparison of La/Sm (ppm) to MgO (wt.%).
Figure A1f. Comparison of Nb/La (ppm) to MgO (wt.%).
90
Figure A1g. Comparison of Ba/TiO2 to MgO (wt.%). Ba is in ppm. TiO2 is in wt.%.
Figure A1h. Comparison of Dy/Yb (ppm) to Mg#.
91
Figure A1i. Comparison of K2O/La to Mg#. K2O is in wt.%. La is in ppm.
Figure A1j. Comparison of K2O/TiO2 (wt.%) to Mg#.
92
Figure A1k. Comparison of La/Sm (ppm) to Mg#.
Figure A1l. Comparison of Nb/La (ppm) to Mg#.
93
Figure A1m. Comparison of Ba/TiO2 to longitude (°). Ba is in ppm. TiO2 is in wt.%.
Figure A1n. Comparison of Dy/Yb (ppm) to longitude (°).
94
Figure A1o. Comparison of K2O/La to longitude (°). K2O is in wt.%. La is in ppm.
Figure A1p. Comparison of K2O/TiO2 (wt.%) to longitude (°).
95
Figure A1q. Comparison of La/Sm (ppm) to longitude (°).
Figure A1r. Comparison of Nb/La (ppm) to longitude (°).
96
8.2) Appendix 2: Global MORB ternary plots
Global MORBs plotted on NKL, TKL, & YSL ternary plots.
Figure A2a. NKL ternary plot showing relative concentrations of Na 2O (wt.% x 5), K2O
(wt.% x 30), & La (ppm) in a variety of basalt glass and whole rock samples. These
include enriched continental magmatic rocks (GEOROC; Lehnert et al., 2000), global
MORBs (PetDB; Lehnert et al., 2000), enriched basalts (Kamenetsky et al., 2000), and
Arctic ridge samples from Lena Trough (Nauret et al., 2011) and Gakkel Ridge (Michael
et al., 2003; Goldstein et al., 2008; Shaw et al., 2010; Snow J.E., in prep; and data from
this study). Also plotted are DMM (Salters & Stracke, 2004) and PM (Sun &
McDonough, 1989).
97
Figure A2b. TKL ternary plot showing relative concentrations of TiO 2 (wt.% x 5), K2O
(wt.% x 30), & La (ppm) in basalt glasses and whole rock samples. Data are from the
same references as those in Figure 12a.
98
Figure A2c. YSL ternary plot showing relative concentrations of Yb (ppm), Sm (ppm x
2), & La (ppm) in basalt glasses and whole rock samples. Data are from the same
references as those in Figure 12a.
99
8.3)
Appendix 3: NbKL & DSL ternary plots
Ternary plots showing relative concentrations of Nb (ppm), K2O (wt. % x 30), &
La (ppm) (NbKL), followed by ternary plots showing relative concentrations of Dy
(ppm), Sm (ppm x 2), & La (ppm) (DSL). Nb is more incompatible than La and K2O.
K2O and La have close incompatibilities, with K2O being slightly more incompatible than
La. Sm is less incompatible than K2O and La.
Data plotted are from this study and from other studies (Sun & McDonough,
1989; Kamenetsky et al., 2000; Goldstein et al., 2003; Michael et al., 2003; Salters &
Stracke, 2004; Standish et al., 2008; Nauret et al., 2011; Shaw et al., 2010; PetDB
(Lehnert et al., 2000); GEOROC (Lehnert et al., 2000)).
Data plotted include: MORB glass samples, continental enriched magma volcanic
whole rock samples, EM1, EM2, and HIMU basalt glass and whole rock samples, global
basalts and enriched continental magmas, and Gakkel Ridge and Lena Trough basalt
glasses.
Ternary plots are:
Fig. A3a: NbKL - global & enriched MORBs
Fig. A3b: NbKL - OIB mantle components
Fig. A3c: NbKL - global & Arctic MORBs
Fig. A3d: DSL - global & enriched MORBs
Fig. A3e: DSL - OIB mantle components
Fig. A3f: DSL - global and Arctic MORBs
100
Figure A3a. NbKL trace element ternary plot: Nb (ppm) & K2O (wt.% x 30) & La
(ppm). Global MORB and enriched basalt glass samples
101
Figure A3b. NbKL trace element ternary plot: Nb (ppm) & K2O (wt.% x 30) & La
(ppm). EM1, EM2, and HIMU basalt glass and whole rock samples
102
Figure A3c. NbKL trace element ternary plot: Nb (ppm) & K2O (wt.% x 30) & La
(ppm). Global basalts and whole rock samples.
103
Figure A3d. DSL trace element ternary plot: Dy (ppm) & Sm (ppm x 2) & La (ppm).
Global MORB glass samples.
104
Figure A3e. DSL trace element ternary plot: Dy (ppm) & Sm (ppm x 2) & La (ppm).
EM1, EM2, and HIMU basalt glass and whole rock samples.
105
Figure A3f. DSL trace element ternary plot: Dy (ppm) & Sm (ppm x 2) & La (ppm).
Global basalt glasses and whole rock samples.
106
8.4)
Appendix 4: NKL ternary plots - mixing lines
NKL ternary plots displaying relative concentrations of Na 2O (wt.% x 5), K2O
(wt.% x 30), & La (ppm) showing two-component mixing lines. K2O and La have close
incompatibilities, with K2O being slightly more incompatible than La. Na 2O is less
incompatible than K2O and La. All diagrams depict the DMM – extreme-K2O CLT
mixing line.
The other mixing lines shown are:
Fig. A4a: NKL - DMM to EM2
Fig. A4b: NKL - DMM to EM1
Fig. A4c: NKL - DMM to average MAR E-MORB
107
Figure A4a. NKL trace element ternary plot showing relative concentrations of Na2O
(wt.% x 5) & K2O (wt.% x 30) & La (ppm). The red two-component mixing line
connects DMM and CLT. The green two-component mixing line connects DMM and
EM2.
108
Figure A4b. NKL trace element ternary plot showing relative concentrations of Na2O
(wt.% x 5) & K2O (wt.% x 30) & La (ppm). The red two-component mixing line
connects DMM and CLT. The blue two-component mixing line connects DMM and
EM1.
109
Figure A4c. NKL trace element ternary plot showing relative concentrations of Na2O
(wt.% x 5) & K2O (wt.% x 30) & La (ppm). The red two-component mixing line
connects DMM and CLT. The dark blue two-component mixing line connects DMM and
MAR E-MORB.
110
8.5)
Appendix 5: Ternary plots - enriched continental magmas
NKL, TKL, YSL, NbKL, and DSL ternary plots showing the enriched continental
magmas: shoshonite, carbonatite, phonolite, and adakite from the GEOROC database
(Lehnert et al., 2000). Rock types plotted are adakite, carbonatite, phonolite, and
shoshonite. Also plotted are depleted MORB mantle (Salters & Stracke, 2004) and
primitive mantle (Sun & McDonough, 1989). The global MORB array, comprising data
from the MAR, EPR, SWIR, & SEIR, (PetDB; Lehnert et al., 2000) is encircled in gray.
Ternary plots are:
Fig. A5a: NKL - Na2O (wt.% x 5), K2O (wt.% x 30), & La (ppm)
Fig. A5b: TKL - TiO2 (wt.% x 5), K2O (wt.% x 30), & La (ppm)
Fig. A5c: YSL - Yb (ppm), Sm (ppm x 2), & La (ppm)
Fig. A5d: NbKL - Nb (ppm), K2O (wt. % x 30), & La (ppm)
Fig. A5e: DSL - Dy (ppm), Sm (ppm x 2), & La (ppm)
111
Figure A5a. NKL ternary plot showing relative concentrations of Na 2O (wt.% x 5), K2O
(wt.% x 30), & La (ppm) in continental volcanic whole rock samples.
112
Figure A5b. TKL ternary plot showing TiO2 (wt.% x 5), K2O (wt.% x 30), & La (ppm)
in continental volcanic whole rock samples.
113
Figure A5c. YSL ternary plot showing relative concentrations of Yb (ppm), Sm (ppm x
2), & La (ppm) in continental volcanic whole rock samples.
114
Figure A5d. NbKl trace element ternary plot: Nb (ppm) & K2O (wt.% x 30) & La (ppm).
Continental enriched magma volcanic whole rock samples
115
Figure A5e. DSL trace element ternary plot: Dy (ppm) & Sm (ppm x 2) & La (ppm).
Continental volcanic whole rock samples
116
8.6)
Appendix 6: NKL & TKL ternary plots - Gakkel Ridge volcanic centers
NKL (Na2O (wt.% x 5), K2O (wt.% x 30), La (ppm)) & TKL (TiO2 (wt.% x 5),
K2O (wt.% x 30), La (ppm)) ternary plots showing the uncertainty polygons surrounding
each plotted sample. Uncertainties were calculated from the standard deviations of the
analyses of the basalt standard VG-2 for the elements Na2O and K2O from electron
microprobe analyses and the basalt standard BIR-1 for the elements La and TiO2 from
LA-ICP-MS analyses. Each plot shows a different section of Gakkel Ridge, separated by
volcanic centers. The global MORB array, comprising data from the MAR, EPR, SWIR,
& SEIR, (PetDB; Lehnert et al., 2000) is encircled in gray. The gray blob represents all
Gakkel Ridge samples from this study and earlier studies (Michael et al., 2003; Goldstein
et al., 2008; Shaw et al., 2010; Snow J.E., in prep).
Ternary plots are:
Fig. A6a: NKL - SMZ13
Fig. A6b: NKL - SMZ19
Fig. A6c: NKL - EVZ31-EVZ37-EVZ-43
Fig. A6d: NKL - EVZ55
Fig. A6e: TKL - SMZ13
Fig. A6f: TKL - SMZ19
Fig. A6g: TKL - EVZ31-EVZ37-EVZ43
Fig. A6h: TKL - EVZ55
117
Figure A6a. NKL ternary plot showing SMZ13 samples and their uncertainty polygons.
118
Figure A6b. NKL ternary plot showing SMZ19 samples and their uncertainty polygons.
119
Figure A6c. NKL ternary plot showing EVZ31, EVZ37, & EVZ43 samples and their
uncertainty polygons.
120
Figure A6d. NKL ternary plot showing EVZ55 samples and their uncertainty polygons.
121
Figure A6e. TKL ternary plot showing SMZ13 samples and their uncertainty polygons.
122
Figure A6f. TKL ternary plot showing SMZ19 samples and their uncertainty polygons.
123
Figure A6g. TKL ternary plot showing EVZ31, EVZ37, & EVZ43 samples and their
uncertainty polygons.
124
Figure A6h. TKL ternary plot showing EVZ55 samples and their uncertainty polygons.
125
8.7) Appendix 7: Tables - basalt standards
The basalt standard VG-2 was used as the independent check standard during
electron microprobe analyses of the major elements of the Gakkel Ridge samples of this
study. The basalt standard BIR-1 was used as the independed check standard during LAICP-MS analyses of the trace elements and the major element TiO2 of the Gakkel Ridge
samples of this study. The basalt standard KL2-G was used as the primary standard
during the LA-ICP-MS analyses of the trace elements and the major element TiO2 of the
Gakkel Ridge samples of this study.
Multiple measurements of each basalt standard were averaged together into
unknowns. The column N is the number of measurements averaged into each unknown.
The “Date” column shows the date on which the measurements were made for each
unknown along with a number for each unknown when there are multiple unknowns for a
single date.
Accepted values of VG-2 are from Jarosewich (1980). Accepted values of KL2-G
and BIR-1 are from (GeoReM; Jochum et al., 2005).
Elemental abundances of zero were not included in the average and standard
deviation calculations.
Tables shown are:
Table A7a: Major element analyses of VG-2
Table A7b: Trace element analyses of BIR-1 (part 1)
Table A7c: TiO2 and trace element analyses of BIR-1 (part 2)
Table A7d: Trace element analyses of KL2-G (part 1)
Table A7e: TiO2 and trace element analyses of KL2-G (part 2)
126
VG-2
Date
20050718 #01
20050718 #02
20050720 #01
20050720 #02
20050720 #03
20050720 #04
20050720 #05
20050720 #06
20050720 #07
20050720 #08
20050720 #09
20050721 #01
20050721 #02
Na 2O
2.83
2.82
2.80
2.77
2.83
2.85
2.89
2.86
2.83
2.77
2.78
2.81
2.84
Average
Standard Deviation
Relative Standard Deviation
Accepted Values
2.82
0.04
1%
2.62
SiO2
50.9
50.9
50.7
50.8
51.0
51.1
50.9
50.9
51.1
50.9
50.8
50.9
50.8
K2O
0.188
0.201
0.191
0.192
0.195
0.193
0.191
0.190
0.197
0.192
0.213
0.193
0.187
FeO
12.0
12.0
11.9
11.9
11.9
11.9
12.0
11.9
12.0
12.0
11.8
11.9
12.0
TiO2
1.88
1.87
1.88
1.88
1.90
1.88
1.89
1.87
1.88
1.88
1.86
1.87
1.86
Al 2O3
14.0
14.1
14.1
14.0
14.1
14.0
14.0
14.0
14.2
14.1
14.0
14.0
14.0
MgO
6.65
6.68
6.68
6.68
6.70
6.66
6.69
6.68
6.70
6.67
6.59
6.66
6.70
CaO
11.1
11.1
11.0
11.1
11.0
11.2
11.0
11.0
11.1
11.1
11.1
11.1
11.1
MnO
0.22
0.20
0.21
0.18
0.20
0.22
0.22
0.19
0.23
0.20
0.22
0.20
0.20
Cl
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
S
0.13
0.14
0.14
0.13
0.14
0.14
0.13
0.13
0.14
0.15
0.14
0.14
0.13
50.9 0.194
0.1 0.01
0.0
4%
50.81 0.19
11.9
0.06
1%
12.1
1.88
0.01
1%
1.85
14.0
0.06
0%
14.06
6.67 11.1
0.03 0.05
0%
0%
6.71 11.12
0.21
0.01
7%
0.22
0.03
0.00
6%
0.14
0.01
4%
P2O5
0.22
0.22
0.20
0.22
0.22
0.23
0.23
0.23
0.23
0.24
0.23
0.21
0.21
Total
100.00
100.25
99.81
99.89
100.25
100.39
100.27
99.94
100.55
100.15
99.74
100.00
100.02
N
4
5
5
4
4
2
4
5
3
3
3
3
6
0.22 100.10
0.01
0.24
5%
0%
0.2 99.86
Table A7a. Major element analyses of the basalt standard VG-2.
BIR-1 (part 1)
Date
Rb
Sr
Y
Zr
Nb
20130309 #01
0.2 115 15.9
15 0.54
20130309 #02
0.2 110 15.6
14 0.55
20130309 #03
0.2 116 16.1
15 0.56
20130309 #04
0.2 113 15.3
14 0.55
20130309 #05
0.2 114 15.5
15 0.56
20130309 #06
0.3 109 15.0
14 0.52
20130309 #07
0.2 114 15.4
15 0.57
Average
Standard Deviation
Relative Standard Deviation
Accepted Values
0.2
0.04
17%
0.2
113
3
2%
109
15.5
0.4
2%
15.6
15
0.4
3%
14
Cs
Ba
La
Ce
Pr
Nd
Sm Eu
Gd
0.039
6.99 0.644
2.07
0.38 2.55 1.11 0.57
1.96
0.049
6.37 0.599
1.97
0.37 2.36 1.14 0.54
1.77
0.050
6.85 0.619
1.96
0.38 2.45 1.20 0.56
1.92
0.025
6.65 0.601
1.92
0.38 2.49 1.08 0.53
1.90
0.024
6.91 0.626
2.06
0.40 2.52 1.13 0.54
1.97
0.022
6.35 0.567
1.87
0.36 2.28 1.05 0.51
1.75
0.026
6.91 0.625
2.00
0.38 2.48 1.14 0.54
1.97
0.55 0.033
0.02 0.012
3% 36%
0.55 0.007
6.72
0.27
4%
7.14
0.612
0.025
4%
0.615
1.98
0.07
4%
1.92
0.38
0.01
3%
0.37
2.45
0.09
4%
2.38
1.12
0.05
4%
1.12
0.54
0.02
4%
0.53
N
6
4
4
5
5
5
5
1.89
0.09
5%
1.87
Table A7b. Trace element analyses of the basalt standard BIR-1.
BIR-1 (part 2)
Date
Tb
Dy
Ho
Er
Tm
Yb
Lu
20130309 #01 1.97 2.73 0.60 1.76 0.26 1.88 0.26
20130309 #02 1.80 2.60 0.55 1.71 0.26 1.78 0.24
20130309 #03 1.80 2.69 0.60 1.79 0.25 1.77 0.24
20130309 #04 1.86 2.58 0.58 1.73 0.24 1.70 0.25
20130309 #05 1.92 2.65 0.62 1.81 0.26 1.79 0.27
20130309 #06 1.75 2.42 0.53 1.62 0.23 1.66 0.24
20130309 #07 1.86 2.64 0.57 1.74 0.26 1.76 0.26
Average
Standard Deviation
Relative Standard Deviation
Accepted Values*
1.85
0.08
4%
0.36
2.62
0.10
4%
2.51
0.58
0.03
5%
0.56
1.74
0.06
4%
1.66
0.25
0.01
5%
0.25
1.76
0.07
4%
1.65
Hf
0.594
0.577
0.585
0.552
0.597
0.527
0.559
Ta
Pb
Th
U
0.0427 4.05238 0.038 0.02
0.0371 3.85858 0.034 0.03
0.0522 3.95675 0.055 0.03
0.0386 3.96506 0.029 0.01
0.0418 4.33158 0.041 0.03
0.0316 3.67548 0.031 0.02
0.0363 3.92934 0.034 0.01
0.25 0.570 0.0400
0.01 0.025 0.0065
5%
4%
16%
0.25 0.582 0.0357
3.96702
0.19936
5%
2.87047
0.037
0.009
23%
0.032
Table A7c. TiO2 and trace element analyses of the basalt standard BIR-1.
127
0.02
0.01
30%
0.01
TiO2 Ni(60)
1.01
197
1.01
191
1.01
194
1.01
192
1.00
194
1.01
188
1.01
193
1.01
0.004
0.4%
0.93
193
3
2%
166
N
6
4
4
5
5
5
5
KL2-G (part 1)
Date
Rb
Sr
Y
Zr
Nb
Cs
Ba
20100707 #01
8.7 356 25.4
152
15 0.115
123
20110209 #01
8.6 348 24.6
148
15 0.112
119
20110209 #02
8.8 364 26.2
156
15 0.118
127
20110209 #03
8.7 353 25.2
151
15 0.113
120
20110209 #04
8.7 358 25.6
153
15 0.117
125
20110209 #05
8.6 353 25.0
150
15 0.116
121
20110209 #06
8.8 360 25.9
154
15 0.114
125
20110209 #07
0
0 25.4
153
15 0.115
122
20110209 #08
0
0 25.4
151
15 0.115
125
20110209 #09
8.6 352 24.8
148
15 0.113
122
20110209 #10
8.8 361 26.1
157
15 0.118
124
20110304 #01
8.5 348 24.6
148
15 0.111
119
20110304 #02
8.8 364 26.2
156
15 0.119
127
20110304 #03
8.7 353 25.0
149
15 0.113
120
20110304 #04
8.7 359 25.4
152
15 0.116
125
20110304 #05
8.6 351 25.0
150
15 0.115
121
20110304 #06
8.8 358 25.9
154
15 0.115
125
20110304 #07
8.7 354 25.0
151
15 0.114
119
20110304 #08
8.7 359 25.5
154
15 0.121
127
20110304 #09
8.6 352 24.5
148
15 0.112
122
20110304 #10
8.8 361 25.8
157
15 0.119
124
20120309 #01
8.1 341 24.0
140
14 0.090
109
20120309 #02
9.0 356 25.4
151
15 0.124
124
20120309 #03
9.1 363 26.1
153
15 0.128
131
20120309 #04
7.9 327 23.3
135
14 0.106
103
20120309 #05
8.9 351 25.9
152
15 0.124
122
20120309 #06
8.4 336 24.0
143
14 0.103
107
20120309 #07
9.1 356 25.4
153
15 0.131
123
20120309 #08
8.3 338 24.0
145
14 0.103
114
20120309 #09
8.6 356 25.5
152
15 0.110
122
20120309 #10
9.1 365 26.1
155
16 0.132
128
20120309 #11
8.0 316 22.3
134
13 0.092
100
20120309 #12
8.6 354 25.5
153
15 0.122
123
20120309 #13
8.7 356 25.5
151
15 0.109
122
20120309 #14
9.0 366 25.6
154
15 0.142
127
Average
Standard Deviation
Relative Standard Deviation
Accepted Values
8.7 353
0.28 10.88
3%
3%
8.7 356
25.2
0.85
3%
25.4
150
5.53
4%
152
15 0.115
0.47 0.01
3%
9%
15 0.115
La
Ce
Pr
Nd
Sm Eu
Gd
13.1 32.9
4.6 21.7 5.55 1.92
5.92
12.6 31.9
4.4 21.0 5.34 1.86
5.66
13.6 33.9
4.8 22.4 5.76 1.98
6.18
12.8 32.1
4.5 21.1 5.34 1.86
5.82
13.4 33.6
4.7 22.2 5.72 1.97
6.00
12.8 32.6
4.5 21.3 5.45 1.88
5.79
13.4 33.3
4.7 22.2 5.68 1.97
6.08
12.9 32.5
4.5
0 5.48 1.89
5.81
13.3 33.4
4.7
0 5.64 1.96
6.05
12.8 32.7
4.5 21.2 5.37 1.89
5.73
13.5 33.2
4.7 22.3 5.78 1.96
6.15
12.6 31.9
4.4 21.0 5.34 1.86
5.66
13.6 33.9
4.8 22.4 5.76 1.98
6.18
12.7 32.1
4.5 21.1 5.34 1.86
5.74
13.4 33.6
4.7 22.2 5.72 1.97
5.93
12.8 32.6
4.5 21.3 5.45 1.88
5.79
13.4 33.3
4.7 22.2 5.68 1.97
6.08
12.5 32.1
4.4 21.1 5.39 1.86
5.63
13.4 34.0
4.7 22.4 5.75 1.99
6.04
12.8 32.7
4.5 21.2 5.37 1.89
5.73
13.5 33.2
4.7 22.3 5.78 1.95
6.15
11.8 29.5
4.1 19.5 4.86 1.70
5.19
13.5 33.7
4.7 22.4 5.73 2.00
6.17
14.3 36.0
5.0 23.5 6.13 2.08
6.47
11.1 28.0
3.9 18.3 4.71 1.66
4.89
13.0 32.9
4.6 21.6 5.52 1.98
5.93
11.5 28.7
4.0 18.8 4.81 1.67
5.08
13.0 33.0
4.6 22.1 5.53 2.01
6.00
12.1 30.6
4.3 20.1 5.19 1.77
5.44
13.1 32.8
4.7 22.0 5.46 1.93
6.10
13.6 34.1
4.9 23.1 5.81 2.06
6.22
10.6 26.5
3.7 17.3 4.46 1.56
4.73
13.0 32.6
4.6 21.3 5.35 1.89
5.92
13.0 33.0
4.6 22.0 5.64 1.92
5.87
13.7 34.1
4.7 22.6 5.83 1.99
6.14
121 12.9
6.91 0.75
6% 6%
123 13.1
32.5
1.84
6%
32.4
4.5
0.26
6%
4.6
Table A7d. Trace element analyses of the basalt standard KL2-G.
128
21.4
1.34
6%
21.6
5.48
0.35
6%
5.54
1.90
0.11
6%
1.92
5.84
0.38
7%
5.92
N
3
5
5
5
6
5
4
5
4
5
4
5
5
5
6
5
4
5
4
5
4
6
6
5
4
4
4
3
6
6
6
5
5
5
4
KL2-G (part 2)
Date
Tb
Dy
20100707 #01
0.89
5.22
20110209 #01
0.85
4.94
20110209 #02
0.93
5.50
20110209 #03
0.87
5.08
20110209 #04
0.91
5.34
20110209 #05
0.87
5.08
20110209 #06
0.92
5.39
20110209 #07
0
5.12
20110209 #08
0
5.35
20110209 #09
0.87
5.01
20110209 #10
0.92
5.48
20110304 #01
0.85
4.94
20110304 #02
0.93
5.50
20110304 #03
0.86
5.07
20110304 #04
0.90
5.34
20110304 #05
0.87
5.08
20110304 #06
0.92
5.39
20110304 #07
0.84
4.95
20110304 #08
0.91
5.36
20110304 #09
0.86
5.00
20110304 #10
0.92
5.50
20120309 #01
0.78
4.60
20120309 #02
0.92
5.43
20120309 #03
0.98
5.72
20120309 #04
0.75
4.30
20120309 #05
0.89
5.26
20120309 #06
0.75
4.51
20120309 #07
0.92
5.54
20120309 #08
0.84
4.92
20120309 #09
0.90
5.16
20120309 #10
0.93
5.58
20120309 #11
0.70
4.26
20120309 #12
0.88
5.10
20120309 #13
0.90
5.29
20120309 #14
0.93
5.33
Average
Standard Deviation
Relative Standard Deviation
Accepted Values
0.88
0.06
7%
0.89
Ho
Er
Tm
Yb
Lu
Hf
Ta
Pb
Th
U
TiO2 Ni(60)
0.961
2.54 0.331
2.1 0.285
3.93 0.961
2.07 1.03 0.548
2.56 112
0.913
2.40 0.314
2.0 0.269
3.70 0.918
2.00 0.97 0.534
2.52 112
1.01
2.68 0.348
2.2 0.301
4.16 1.00
2.14 1.09 0.562
2.60 112
0.932
2.48 0.323
2.0 0.281
3.83 0.942
2.04 1.00 0.519
2.55 114
0.986
2.59 0.338
2.1 0.288
4.01 0.977
2.10 1.06 0.572
2.57 110
0.934
2.49 0.319
2.0 0.276
3.77 0.923
2.01 1.00 0.549
2.54 113
1.00
2.60 0.346
2.2 0.296
4.13 1.01
2.15 1.07 0.547
2.59 110
0
2.47 0.325
2.0 0.279
0 0.946
0
0 0.523
0 115
0
2.62 0.338
2.2 0.292
0 0.980
0
0 0.579
0 109
0.924
2.46 0.317
2.0 0.276
3.78 0.925
2.02 1.00 0.528
2.54 111
1.01
2.64 0.349
2.2 0.296
4.12 1.01
2.13 1.07 0.573
2.58 113
0.913
2.40 0.314
2.0 0.269
3.70 0.918
2.00 0.97 0.533
2.48 112
1.01
2.68 0.348
2.2 0.301
4.16 1.00
2.14 1.09 0.563
2.56 112
0.913
2.48 0.320
2.0 0.280
3.82 0.941
2.04 1.00 0.519
2.54 116
0.968
2.59 0.335
2.1 0.289
4.02 0.978
2.10 1.06 0.572
2.56 112
0.934
2.49 0.319
2.0 0.276
3.77 0.923
2.01 1.00 0.549
2.54 113
1.00
2.60 0.346
2.2 0.296
4.13 1.01
2.15 1.07 0.547
2.59 110
0.910
2.39 0.315
2.0 0.270
3.71 0.933
2.06 0.99 0.516
2.57 116
0.985
2.61 0.338
2.2 0.293
4.03 0.997
2.08 1.09 0.588
2.55 110
0.924
2.46 0.317
2.0 0.276
3.77 0.922
2.03 1.00 0.548
2.54 111
1.01
2.64 0.349
2.2 0.297
4.13 1.01
2.13 1.07 0.594
2.58 113
0.843
2.19 0.293
1.8 0.252
3.39 0.846
1.77 0.90 0.463
2.56 106
1.02
2.62 0.344
2.2 0.289
4.01 1.01
2.19 1.07 0.592
2.56 115
1.03
2.71 0.353
2.4 0.317
4.18 1.05
2.29 1.13 0.581
2.57 115
0.812
2.10 0.280
1.9 0.234
3.26 0.815
1.74 0.85 0.497
2.54 109
0.943
2.55 0.328
2.3 0.285
3.91 1.00
2.09 1.02 0.599
2.58 112
0.837
2.20 0.286
1.8 0.243
3.41 0.841
1.88 0.89 0.476
2.56 110
1.01
2.54 0.349
2.1 0.300
3.91 0.947
2.09 1.07 0.644
2.56 114
0.889
2.35 0.300
1.9 0.262
3.68 0.895
1.89 0.94 0.519
2.51 110
0.963
2.62 0.338
2.1 0.292
3.97 0.962
2.11 1.02 0.581
2.56 111
1.03
2.65 0.355
2.2 0.301
4.15 0.990
2.21 1.08 0.545
2.61 115
0.769
2.03 0.260
1.6 0.233
3.07 0.780
1.64 0.83 0.422
2.51 109
0.953
2.47 0.330
2.1 0.277
3.92 0.940
2.02 1.01 0.544
2.58 113
0.958
2.56 0.327
2.0 0.283
3.93 0.957
2.04 1.03 0.539
2.55 112
1.00
2.67 0.349
2.2 0.300
4.07 1.00
2.20 1.06 0.640
2.60 114
5.16 0.948
0.34 0.06
7%
7%
5.22 0.961
2.50 0.327
0.16
0.02
7%
7%
2.54 0.331
2.1
0.15
7%
2.1
0.282
0.02
7%
0.285
3.86 0.950
0.27 0.06
7%
6%
3.93 0.961
2.05
0.14
7%
2.07
1.02
0.07
7%
1.02
0.549
0.04
8%
0.548
Table A7e. TiO2 and tracele element analyses of the basalt standard KL2-G.
129
2.56
0.03
1%
2.56
112
2.28
2%
112
N
3
5
5
5
6
5
4
5
4
5
4
5
5
5
6
5
4
5
4
5
4
6
6
5
4
4
4
3
6
6
6
5
5
5
4
8.8) Appendix 8: Tables - relative uncertainty ranges
Relative uncertainties for each element for each sample were calculated by
dividing the uncertainty of each element (the standard deviation of all measurements of
the standard VG-2 for major elements and BIR-1 for trace elements) by the measured
concentration (ppm) of that element in each sample and multiplying by 100 to result in a
percent value. Percentages were rounded to zero decimal places. Aggregated relative
uncertainty ranges are presented for each Gakkel Ridge volcanic zone in the tables below.
Volcanic centers EVZ31, EVZ37, and EVZ43 are all grouped together as EVZ31-37-43.
Major elements
Zone
Mg#
MgO
SMZ13
<1%
<1%
SMZ19
<1%
<1%
EVZ31-37-43 <1% <1-1%
EVZ55
<1%
<1%
SiO2
<1%
<1%
<1%
<1%
Al 2O3
<1%
<1%
<1%
<1%
Na 2O K2O TiO2
1% 1-2%
1%
1% 1-12%
1%
1% 2-3% <1-1%
1% 2-6%
1%
CaO
<1%
<1%
<1%
<1%
FeO
1%
1%
1%
1%
MnO
Cl
8-10% 4-11%
8-11% 4-20%
7-10% 7-20%
8-10% 12-41%
S
6-8%
5-7%
4-6%
5-6%
P2O5
3-6%
5-14%
3-6%
5-11%
Table A8a. Relative uncertainty ranges of the major elements.
Trace elements (part 1)
Zone
SMZ13
SMZ19
EVZ31-37-43
EVZ55
Rb
Sr
Y
Zr
Nb
Cs
<1-1% 1-2%
1% <1%
<1%
3-12%
<1-6% 1-2% 1-2% <1-1% <1-1%
4-74%
<1-1% 1-2%
1% <1%
<1% 11-25%
1-3% 1-2% 1-2% <1-1% <1-1% 17-34%
Ba
La
Ce
Pr
Nd
Sm
Eu
Gd
<1%
<1%
<1% <1-1%
1%
1% 1-2%
2%
<1-3% <1-1% <1-1% <1-1%
1% 1-2% 1-2% 2-3%
<1-1%
<1% <1-1% <1-1% <1-1% 1-2% 1-2% 1-2%
<1-3% <1-1% <1-1% <1-1% 1-2% 1-2%
2% 2-3%
Table A8b. Relative uncertainty ranges of the trace elements.
Trace elements (part 2)
Zone
Tb
Dy
Ho
Er
Tm
Yb
Lu
Hf
Ta
SMZ13
8-11% 1-2% 2-3% 1-2% 2-3% 2-3% 2-3%
1% 1-2%
SMZ19
9-13% 2-3% 3-4%
2% 2-4% 2-3%
3% 1-2% 1-8%
EVZ31-37-43 6-12% 1-2% 2-3% 1-2% 2-3% 1-3% 1-3%
1% 1-2%
EVZ55
11-14%
2% 3-4%
2% 3-4% 2-3%
3% 1-2% 1-4%
Pb
Th
U
TiO2 Sc
14-30% <1-2%
1-5% <1%
8-9%
20-58% 1-8% 2-19% <1% 7-10%
20-33% 1-2%
2-5% <1%
7-9%
26-58% 1-5% 3-11% <1%
7-9%
Table A8c. Relative uncertainty ranges of trace elements and TiO2.
130
9)
REFERENCES
Brodholt J.P., Batiza R. (1989) Global systematics of unaveraged mid-ocean ridge basalt
compositions: comment on “global correlations of ocean ridge basalt chemistry with
axial depth and crustal thickness” by E.M. Klein and C.H. Langmuir. Journal of
Geophysical Reasearch, vol 94, p 4231-4239.
Cochran J.R. (2008) Seamount volcanism along the Gakkel Ridge, Arctic Ocean.
Geophysics Journal International, vol 174, p 1153-1173.
Collier M.L., Kelemen P.B. (2010) The case for reactive crystallization at mid-ocean
ridges. Journal of Petrology, vol 51, p 1913-1940.
Dick H.J.B., Lin J., Schouten H. (2003) An ultraslow-spreading class of ocean ridge.
Nature, vol 426, p 405-412.
Elthon D., Stewart M., Ross D.K. (1992) Compositional trends of minerals in oceanic
cumulates. Journal of Geophysical Research, vol 97, p 189-199.
Engen O., Eldholm O., Bungum H. (2003) The Arctic plate boundary. Journal of
Geophysical Research, vol 108(B2), 2075
Goldstein S.L., Soffer G., Langmuir C.H., Lehnert K.A., Graham D.W., and Michael P.J.
(2008) Origin of a ‘southern hemisphere’ geochemical signature in the arctic upper
mantle. Nature, vol 453, Issue 7191, p 89-93. DOI:10.1038/nature06919
Graham D.J., Midgley N.G. (2000) Graphical representation of particle shape using
triangular diagrams: an Excel spreadsheet method. Earth Surface Processes and
Landforms, vol 25, p 1473-1477.
Grove T.L., Kinzler R.J., Bryan W.B. (1992) Fractionation of mid-ocean ridge basalt
(MORB). In: Morgan J.P., Blackman D.K., Sinton J.M. (Eds.), Mantle Flow and Melt
Generation at Mid-Ocean Ridges, Geophysical Monograph Series, American
Geophysical Union, Washington, vol 71, p 281-310.
Hart S.R. (1984) A large-scale isotopic anomaly in the Southern Hemisphere mantle.
Nature, vol 309, p 753-757.
Hart S.R., Zindler A. (1986) In search of a bulk-earth composition. Chemical Geology,
vol 57, p 247-267.
Helffrich G.R., Wood B.J. (2001) The Earth’s mantle. Nature, vol 412, p 501-507.
Hofmann A.W. (1988) Chemical differentiation of the Earth: the relationship between
mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters, vol
90, p 297-314.
131
Hofmann A.W. (1997) Mantle geochemistry: the message from oceanic volcanism.
Nature, vol 385, p 219-229.
Hofmann A.W. (2007) The lost continents. Nature, vol 448, p 655-656.
Hofmann A.W., White W.M. (1982) Mantle plumes from ancient oceanic crust. Earth
and Planetary Science Letters, vol 57, p 421-436.
Holness M.B., Richter F.M. (1989) Possible effects of spreading rate on MORB isotopic
and rare earth composition arising. The Journal of Geology, vol 97, p 247-260.
Ionov D.A., Mukasa, S.B., Bodinier J.L. (2002) Sr-Nd-Pb isotopic compositions of
peridotite xenoliths from Spitsbergen: Numerical modeling indicates Sr-Nd
decoupling in the mantle by melt percolation metasomatism. Journal of Petrology, vol
43, p 2261-2278.
Jarosewich, E., Nelen J.A., Norberg J.A. (1980) Reference samples for electron
microprobe analysis. Geostandards Newsletter, vol 4, p 43-47.
Jarosewich, E. (2002) Smithsonian microbeam standards. Journal of Research of the
National Institute of Standards and Technology, vol 107, p 681-685.
Jochum K.P., Nohl U., Herwig K., Lammel E., Stoll B., Hofmann, A.W. (2005)
GeoReM: A new geochemical database for reference materials and isotopic
standards. Geostandards and Geoanalytical Research, vol 29, p 333-338.
Jokat W., Ritzmann O., Schmidt-Aursch M.C., Drachev S., Gauger S., Snow J. (2003)
Geophysical evidence for reduced melt production on the Arctic ultraslow Gakkel
mid-ocean ridge. Nature, vol 423, p 462-465.
Kamenetsky V.S., Everard J.L., Crawford A.J., Varne R., Eggins, S.M., Layton R. (2000)
Enriched end-member of primitive MORB melts: Petrology and geochemistry of
glasses from Macquarie Island (SW Pacific). Journal of Petrology, vol 41, p 411-430.
Kay R., Hubbard N.J., Gast P.W. (1970) Chemical characteristics and origin of oceanic
ridge volcanic rocks. Journal of Geophysical Research, vol 75, p 1585-1613.
Kelemen P.B., Hirth G., Shimizu N., Spiegelman M., Dick H.J.B. (1997) A review of melt
migration processes in the adiabatically upwelling mantle beneath oceanic spreading
ridges. Philosophical Transactions of the Royal Society of London A, vol 355, p 283318.
Klein E.M. (2003a) Spread thin in the arctic. Nature, vol 423, Issue 6943, p 932-933.
DOI:10.1038/423932a
Klein E.M., (2003b) Geochemistry of the igneous ocean crust. In: Rudnick, R.L. (Ed.),
The Crust: Treatise on Geochemistry, Elsevier-Pergamon, Oxford, vol 3, p 443-463.
132
Langmuir C.H., Klein E.M., Plank T. (1992) Petrological systematics of mid-ocean ridge
basalts: constraints on melt generation beneath ocean ridges. In: Morgan J.P.,
Blackman D.K., Sinton J.M. (Eds.), Mantle Flow and Melt Generation at Mid-Ocean
Ridges, Geophysical Monograph Series, American Geophysical Union, Washington
D.C., vol 71, p 183-280.
Langmuir C.H., Forsyth D.W. (2007) Mantle melting beneath mid-ocean ridges.
Oceanography, vol 20, p 78-89.
Lehnert K., Su Y., Langmuir C.H., Sarbas B., Nohl U. (2000) A global geochemical
database structure for rocks. Geochemistry Geophysics Geosystems, vol 1,
DOI:10.1029/1999GC000026.
Le Roex A.P., Dick H.J.B., Watkins R.T. (1992) Petrogenesis of anomalous K-enriched
MORB from the Southwest Indian Ridge: 11°53’E to 14°38’E. Contributions to
Mineralogy and Petrology, vol 110, p 253-268.
Liu C.Z., Snow J.E., Hellebrand E., Brugmann G., Handt A.V.D., Buchl A., Hofmann
A.W. (2008) Ancient, highly heterogeneous mantle beneath Gakkel ridge, Arctic
Ocean. Nature, vol 452, p 311-316.
McDonough W.F., Sun S.S. (1995) The composition of the Earth. Chemical Geology, vol
120, p 223-353.
Michael P.J., Langmuir C.H., Dick H.J.B., Snow J.E., Goldstein S.L., Graham D.W.,
Lehnert K., Kurras G., Jokat W., Mühe R., and Edmonds HN. (2003) Magmatic and
amagmatic seafloor generation at the ultraslow-spreading Gakkel ridge, Arctic
Ocean. Nature, vol 423, Issue 6943, p 956-U1. DOI:10.1038/nature01704
Mühe R.K., Bohrmann H., Hormann P.K., Thiede J., Stoffers P. (1991) Spinifex basalts
with komatiite-tholeitte trend from the Nansen-Gakkel Ridge (Arctic-Ocean).
Tectonophysics, vol 190, p 95-108.
Mühe R., Devey C.W., Bohrmann H. (1993) Isotope and trace element geochemistry of
MORB from the Nansen-Gakkel ridge at 86° north. Earth and Planetary Science
Letters, vol 120, p 103-109.
Mühe R., Bohrmann H., Garbe-Schonberg D., Kassens H. (1997) E-MORB glasses from
the Gakkel Ridge (Arctic Ocean) at 87°N: evidence for the Earth’s most northerly
volcanic activity. Earth and Planetary Science Letters, vol 152, p 1-9.
Nauret F., Moreira M., Snow J.E. (2010) Rare gases in lavas from the ultraslow
spreading Lena Trough, Arctic Ocean. Geochemistry Geophysics Geosystems, vol
11, Q0AC04
Nauret E., Snow J.E., Hellebrand E., Weis D. (2011) Geochemical Composition of K-rich
Lavas from the Lena Trough (Arctic Ocean). Journal of Petrology, vol 52, p 11851206.
133
Navon O., Stolper E. (1987) Geochemical consequences of melt percolation: the upper
mantle as a chromatographic column. The Journal of Geology, vol 95, p 285-307.
Ohara M.J. (1965) Are ocean floor basalts primary magma? Nature, vol 220, p 683-686.
Salters V.J.M., Stracke A. (2004) Composition of the depleted mantle. Geochemistry
Geophysics Geosystems, vol 5, Q05004.
Schilling J.G. (1971) Sea-floor evolution: rare-earth evidence. Philosophical
Transactions of the Royal Society of London. Series A, Mathematical and Physical
Sciences, vol 268, p 663-706.
Schilling J.G. (1973) Iceland mantle plume: Geochemical study of Reykjanes Ridge.
Nature, vol 242, p 565-571.
Shaw A.M., Behn M.D., Humphris S.E., Sohn R.A., Gregg P.M. (2010) Deep pooling of
low degree melts and volatile fluxes at the 85E segment of Gakkel Ridge: Evidence
from olivine-hosted melt inclusions and glasses. Earth and Planetary Science Letters,
vol 289, p 311-322.
Snow J.E., Jokat W., Hellebrand E., Mühe R. (2001) Magmatic and hydrothermal activity
in Lena Trough, Arctic Ocean. EOS, Transactions American Geophysical Union, vol
82, p 193, 197-198.
Snow, J.E. and Petrology Group, ARK VXII-2 (2002) Petrogenesis of crustal rocks. In:
Thiede, J. et al. (2002) Cruise report, ARK XVII/2. Reports on Polar and Marine
Research, vol 421, p 118-164.
Snow J.E., Edmonds H.M. (2007) Ultraslow-spreading ridges rapid paradigm changes.
Oceanography, vol 20, p 90-101.
Snow J.E., Hellebrand E., Handt A.V.D., Nauret F., Gao Y., Schenke H.W. (2011)
Oblique nonvolcanic seafloor spreading in Lena Trough Arctic Ocean. Geochemistry
Geophysics Geosystems, Vol 12, Q10009.
Sobolev A.V., Hofmann A.W., Sobolev S.V., Nikogosian I.K. (2005) An olivine-free
mantle source of Hawaiian shield basalts. Nature, vol 434, p 590-597.
Sobolev A.V, Hofmann A.W., Kuzmin D.V., Yaxley G.M., Arndt N.T., Chung S.L.,
Danyushevsky L.V., Elliot T., Frey F.A., Garcia M.O., Gurenko A.A., Kamenetsky
V.S., Kerr A.C., Krivolutskaya N.A., Matvienkov V.V., Nikogosian I.K., Rocholl A.,
Sigurdsson I.A., Sushchevskaya N.M., Teklay M. (2007) The amount of recycled
crust in sources of mantle-derived melts. Science, vol 316, p 412-417.
Standish J.J., Dick H.J.B., Michael P.J., Melson W.G., O’Hearn T. (2008) MORB
generation beneath the ultraslow spreading Southwest Indian Ridge (9-25E): Major
element chemistry and the importance of process versus source. Geochemistry
Geophysics Geosystems, vol 9, Q05004.
134
Stolper E. (1980) A phase diagram for mid-ocean ridge basalts: preliminary results and
implications for petrogenesis. Contributions to Mineralogy and Petrology, vol 74, p
13-27.
Sun S.S., McDonough W.F. (1989) Chemical and isotopic systematics of oceanic basalts:
implications for mantle composition and processes. Geological Society, London,
Special Publications, vol 42, p 313-345.
Van Keken P.E., Hauri E.H., Ballentine C.J. (2002) Mantle mixing: generation,
preservation, and destruction of chemical heterogeneity. Annual Review of Earth and
Planetary Sciences, vol 30, p 493-525
Walker D., Shibata T., DeLong S.E. (1979) Abyssal tholeiites from the Oceanographer
Fracture Zone. Contributions to Mineralogy and Petrology, vol 70, p 111-125.
Weaver B.L. (1991a) The origin of ocean island basalt end-member compositions: Trace
element and isotopic constraints. Earth and Planetary Science Letters, vol 104, p 381397.
Weaver B.L. (1991b) Trace element evidence for the origin of ocean-island basalts.
Geology, vol 19, p 123-126.
White, W.M. (2013) Geochemistry, 1st edition. West Sussex: Wiley-Blackwell.
Winter, J.D. (2001) An Introduction to Igneous and Metamorphic Petrology. PrenticeHall Inc.
Wood D.A. (1979) A variably veined suboceanic upper mantle – Genetic significance for
mid-ocean ridge basalts from geochemical evidence. Geology, vol 7, p 499-503.
Workman R.K., Hart S.R. (2005) Major and trace element composition of the depleted
MORB mantle (DMM). Earth and Planetary Science Letters, vol 231, p 53-72.
Zindler A., Hart S.R. (1986) Chemical geodynamics. Annual Review of Earth and
Planetary Sciences. vol, 14, p 493-571.
135